Spaces:
Running
on
A10G
Running
on
A10G
2002 Turkish pop, a style of energetic, danceable songs Take artists like Ajda Pekkan, Tarkan, Sertab Erener, Mustafa Sandal as an example and compose a song
#20
by
StarStoner11
- opened
This view is limited to 50 files because it contains too many changes.
See the raw diff here.
- .github/actions/audiocraft_build/action.yml +0 -2
- .github/workflows/audiocraft_docs.yml +3 -3
- .github/workflows/audiocraft_tests.yml +1 -6
- .gitignore +1 -8
- CHANGELOG.md +1 -34
- CONTRIBUTING.md +2 -2
- LICENSE_weights +157 -399
- MANIFEST.in +0 -7
- model_cards/MUSICGEN_MODEL_CARD.md → MODEL_CARD.md +7 -31
- Makefile +4 -23
- README.md +80 -46
- app.py +216 -0
- app_batched.py +222 -0
- assets/a_duck_quacking_as_birds_chirp_and_a_pigeon_cooing.mp3 +0 -0
- assets/sirens_and_a_humming_engine_approach_and_pass.mp3 +0 -0
- audiocraft/__init__.py +1 -17
- audiocraft/adversarial/__init__.py +0 -22
- audiocraft/adversarial/discriminators/__init__.py +0 -10
- audiocraft/adversarial/discriminators/base.py +0 -34
- audiocraft/adversarial/discriminators/mpd.py +0 -106
- audiocraft/adversarial/discriminators/msd.py +0 -126
- audiocraft/adversarial/discriminators/msstftd.py +0 -134
- audiocraft/adversarial/losses.py +0 -228
- audiocraft/data/__init__.py +1 -3
- audiocraft/data/audio.py +21 -37
- audiocraft/data/audio_dataset.py +31 -93
- audiocraft/data/audio_utils.py +10 -12
- audiocraft/data/info_audio_dataset.py +0 -110
- audiocraft/data/music_dataset.py +0 -270
- audiocraft/data/sound_dataset.py +0 -330
- audiocraft/data/zip.py +6 -8
- audiocraft/environment.py +0 -176
- audiocraft/grids/__init__.py +0 -6
- audiocraft/grids/_base_explorers.py +0 -80
- audiocraft/grids/audiogen/__init__.py +0 -6
- audiocraft/grids/audiogen/audiogen_base_16khz.py +0 -23
- audiocraft/grids/audiogen/audiogen_pretrained_16khz_eval.py +0 -68
- audiocraft/grids/compression/__init__.py +0 -6
- audiocraft/grids/compression/_explorers.py +0 -55
- audiocraft/grids/compression/debug.py +0 -31
- audiocraft/grids/compression/encodec_audiogen_16khz.py +0 -29
- audiocraft/grids/compression/encodec_base_24khz.py +0 -28
- audiocraft/grids/compression/encodec_musicgen_32khz.py +0 -34
- audiocraft/grids/diffusion/4_bands_base_32khz.py +0 -27
- audiocraft/grids/diffusion/__init__.py +0 -6
- audiocraft/grids/diffusion/_explorers.py +0 -66
- audiocraft/grids/musicgen/__init__.py +0 -6
- audiocraft/grids/musicgen/_explorers.py +0 -93
- audiocraft/grids/musicgen/musicgen_base_32khz.py +0 -43
- audiocraft/grids/musicgen/musicgen_base_cached_32khz.py +0 -67
.github/actions/audiocraft_build/action.yml
CHANGED
@@ -21,8 +21,6 @@ runs:
|
|
21 |
python3 -m venv env
|
22 |
. env/bin/activate
|
23 |
python -m pip install --upgrade pip
|
24 |
-
pip install torch torchvision torchaudio
|
25 |
-
pip install xformers
|
26 |
pip install -e '.[dev]'
|
27 |
- name: System Dependencies
|
28 |
shell: bash
|
|
|
21 |
python3 -m venv env
|
22 |
. env/bin/activate
|
23 |
python -m pip install --upgrade pip
|
|
|
|
|
24 |
pip install -e '.[dev]'
|
25 |
- name: System Dependencies
|
26 |
shell: bash
|
.github/workflows/audiocraft_docs.yml
CHANGED
@@ -23,9 +23,9 @@ jobs:
|
|
23 |
- name: Make docs
|
24 |
run: |
|
25 |
. env/bin/activate
|
26 |
-
make
|
27 |
-
git add -f
|
28 |
-
git commit -m
|
29 |
|
30 |
- name: Push branch
|
31 |
run: |
|
|
|
23 |
- name: Make docs
|
24 |
run: |
|
25 |
. env/bin/activate
|
26 |
+
make docs
|
27 |
+
git add -f docs
|
28 |
+
git commit -m docs
|
29 |
|
30 |
- name: Push branch
|
31 |
run: |
|
.github/workflows/audiocraft_tests.yml
CHANGED
@@ -12,11 +12,6 @@ jobs:
|
|
12 |
steps:
|
13 |
- uses: actions/checkout@v2
|
14 |
- uses: ./.github/actions/audiocraft_build
|
15 |
-
-
|
16 |
-
run: |
|
17 |
. env/bin/activate
|
18 |
make tests
|
19 |
-
- name: Run integration tests
|
20 |
-
run: |
|
21 |
-
. env/bin/activate
|
22 |
-
make tests_integ
|
|
|
12 |
steps:
|
13 |
- uses: actions/checkout@v2
|
14 |
- uses: ./.github/actions/audiocraft_build
|
15 |
+
- run: |
|
|
|
16 |
. env/bin/activate
|
17 |
make tests
|
|
|
|
|
|
|
|
.gitignore
CHANGED
@@ -35,7 +35,7 @@ wheels/
|
|
35 |
.coverage
|
36 |
|
37 |
# docs
|
38 |
-
/
|
39 |
|
40 |
# dotenv
|
41 |
.env
|
@@ -46,13 +46,6 @@ wheels/
|
|
46 |
venv/
|
47 |
ENV/
|
48 |
|
49 |
-
# egs with manifest files
|
50 |
-
egs/*
|
51 |
-
!egs/example
|
52 |
-
# local datasets
|
53 |
-
dataset/*
|
54 |
-
!dataset/example
|
55 |
-
|
56 |
# personal notebooks & scripts
|
57 |
*/local_scripts
|
58 |
*/notes
|
|
|
35 |
.coverage
|
36 |
|
37 |
# docs
|
38 |
+
/docs
|
39 |
|
40 |
# dotenv
|
41 |
.env
|
|
|
46 |
venv/
|
47 |
ENV/
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
# personal notebooks & scripts
|
50 |
*/local_scripts
|
51 |
*/notes
|
CHANGELOG.md
CHANGED
@@ -4,37 +4,7 @@ All notable changes to this project will be documented in this file.
|
|
4 |
|
5 |
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
|
6 |
|
7 |
-
## [
|
8 |
-
|
9 |
-
Adding stereo models.
|
10 |
-
|
11 |
-
|
12 |
-
## [1.1.0] - 2023-11-06
|
13 |
-
|
14 |
-
Not using torchaudio anymore when writing audio files, relying instead directly on the commandline ffmpeg. Also not using it anymore for reading audio files, for similar reasons.
|
15 |
-
|
16 |
-
Fixed DAC support with non default number of codebooks.
|
17 |
-
|
18 |
-
Fixed bug when `two_step_cfg` was overriden when calling `generate()`.
|
19 |
-
|
20 |
-
Fixed samples being always prompted with audio, rather than having both prompted and unprompted.
|
21 |
-
|
22 |
-
**Backward incompatible change:** A `torch.no_grad` around the computation of the conditioning made its way in the public release.
|
23 |
-
The released models were trained without this. Those impact linear layers applied to the output of the T5 or melody conditioners.
|
24 |
-
We removed it, so you might need to retrain models.
|
25 |
-
|
26 |
-
**Backward incompatible change:** Fixing wrong sample rate in CLAP (WARNING if you trained model with CLAP before).
|
27 |
-
|
28 |
-
**Backward incompatible change:** Renamed VALLEPattern to CoarseFirstPattern, as it was wrongly named. Probably no one
|
29 |
-
retrained a model with this pattern, so hopefully this won't impact you!
|
30 |
-
|
31 |
-
|
32 |
-
## [1.0.0] - 2023-09-07
|
33 |
-
|
34 |
-
Major revision, added training code for EnCodec, AudioGen, MusicGen, and MultiBandDiffusion.
|
35 |
-
Added pretrained model for AudioGen and MultiBandDiffusion.
|
36 |
-
|
37 |
-
## [0.0.2] - 2023-08-01
|
38 |
|
39 |
Improved demo, fixed top p (thanks @jnordberg).
|
40 |
|
@@ -45,9 +15,6 @@ More options when launching Gradio app locally (thanks @ashleykleynhans).
|
|
45 |
|
46 |
Testing out PyTorch 2.0 memory efficient attention.
|
47 |
|
48 |
-
Added extended generation (infinite length) by slowly moving the windows.
|
49 |
-
Note that other implementations exist: https://github.com/camenduru/MusicGen-colab.
|
50 |
-
|
51 |
## [0.0.1] - 2023-06-09
|
52 |
|
53 |
Initial release, with model evaluation only.
|
|
|
4 |
|
5 |
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
|
6 |
|
7 |
+
## [0.0.2a] - TBD
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
Improved demo, fixed top p (thanks @jnordberg).
|
10 |
|
|
|
15 |
|
16 |
Testing out PyTorch 2.0 memory efficient attention.
|
17 |
|
|
|
|
|
|
|
18 |
## [0.0.1] - 2023-06-09
|
19 |
|
20 |
Initial release, with model evaluation only.
|
CONTRIBUTING.md
CHANGED
@@ -1,11 +1,11 @@
|
|
1 |
-
# Contributing to
|
2 |
|
3 |
We want to make contributing to this project as easy and transparent as
|
4 |
possible.
|
5 |
|
6 |
## Pull Requests
|
7 |
|
8 |
-
|
9 |
Therefore, we do not plan on accepting many pull requests for new features.
|
10 |
We certainly welcome them for bug fixes.
|
11 |
|
|
|
1 |
+
# Contributing to Audiocraft
|
2 |
|
3 |
We want to make contributing to this project as easy and transparent as
|
4 |
possible.
|
5 |
|
6 |
## Pull Requests
|
7 |
|
8 |
+
Audiocraft is the implementation of a research paper.
|
9 |
Therefore, we do not plan on accepting many pull requests for new features.
|
10 |
We certainly welcome them for bug fixes.
|
11 |
|
LICENSE_weights
CHANGED
@@ -1,399 +1,157 @@
|
|
1 |
-
Attribution-NonCommercial 4.0 International
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
Creative Commons public licenses
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
terms and conditions,
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
101 |
-
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
|
114 |
-
|
115 |
-
|
116 |
-
|
117 |
-
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
|
125 |
-
|
126 |
-
|
127 |
-
|
128 |
-
|
129 |
-
|
130 |
-
|
131 |
-
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
its terms and conditions.
|
159 |
-
|
160 |
-
3. Term. The term of this Public License is specified in Section
|
161 |
-
6(a).
|
162 |
-
|
163 |
-
4. Media and formats; technical modifications allowed. The
|
164 |
-
Licensor authorizes You to exercise the Licensed Rights in
|
165 |
-
all media and formats whether now known or hereafter created,
|
166 |
-
and to make technical modifications necessary to do so. The
|
167 |
-
Licensor waives and/or agrees not to assert any right or
|
168 |
-
authority to forbid You from making technical modifications
|
169 |
-
necessary to exercise the Licensed Rights, including
|
170 |
-
technical modifications necessary to circumvent Effective
|
171 |
-
Technological Measures. For purposes of this Public License,
|
172 |
-
simply making modifications authorized by this Section 2(a)
|
173 |
-
(4) never produces Adapted Material.
|
174 |
-
|
175 |
-
5. Downstream recipients.
|
176 |
-
|
177 |
-
a. Offer from the Licensor -- Licensed Material. Every
|
178 |
-
recipient of the Licensed Material automatically
|
179 |
-
receives an offer from the Licensor to exercise the
|
180 |
-
Licensed Rights under the terms and conditions of this
|
181 |
-
Public License.
|
182 |
-
|
183 |
-
b. No downstream restrictions. You may not offer or impose
|
184 |
-
any additional or different terms or conditions on, or
|
185 |
-
apply any Effective Technological Measures to, the
|
186 |
-
Licensed Material if doing so restricts exercise of the
|
187 |
-
Licensed Rights by any recipient of the Licensed
|
188 |
-
Material.
|
189 |
-
|
190 |
-
6. No endorsement. Nothing in this Public License constitutes or
|
191 |
-
may be construed as permission to assert or imply that You
|
192 |
-
are, or that Your use of the Licensed Material is, connected
|
193 |
-
with, or sponsored, endorsed, or granted official status by,
|
194 |
-
the Licensor or others designated to receive attribution as
|
195 |
-
provided in Section 3(a)(1)(A)(i).
|
196 |
-
|
197 |
-
b. Other rights.
|
198 |
-
|
199 |
-
1. Moral rights, such as the right of integrity, are not
|
200 |
-
licensed under this Public License, nor are publicity,
|
201 |
-
privacy, and/or other similar personality rights; however, to
|
202 |
-
the extent possible, the Licensor waives and/or agrees not to
|
203 |
-
assert any such rights held by the Licensor to the limited
|
204 |
-
extent necessary to allow You to exercise the Licensed
|
205 |
-
Rights, but not otherwise.
|
206 |
-
|
207 |
-
2. Patent and trademark rights are not licensed under this
|
208 |
-
Public License.
|
209 |
-
|
210 |
-
3. To the extent possible, the Licensor waives any right to
|
211 |
-
collect royalties from You for the exercise of the Licensed
|
212 |
-
Rights, whether directly or through a collecting society
|
213 |
-
under any voluntary or waivable statutory or compulsory
|
214 |
-
licensing scheme. In all other cases the Licensor expressly
|
215 |
-
reserves any right to collect such royalties, including when
|
216 |
-
the Licensed Material is used other than for NonCommercial
|
217 |
-
purposes.
|
218 |
-
|
219 |
-
Section 3 -- License Conditions.
|
220 |
-
|
221 |
-
Your exercise of the Licensed Rights is expressly made subject to the
|
222 |
-
following conditions.
|
223 |
-
|
224 |
-
a. Attribution.
|
225 |
-
|
226 |
-
1. If You Share the Licensed Material (including in modified
|
227 |
-
form), You must:
|
228 |
-
|
229 |
-
a. retain the following if it is supplied by the Licensor
|
230 |
-
with the Licensed Material:
|
231 |
-
|
232 |
-
i. identification of the creator(s) of the Licensed
|
233 |
-
Material and any others designated to receive
|
234 |
-
attribution, in any reasonable manner requested by
|
235 |
-
the Licensor (including by pseudonym if
|
236 |
-
designated);
|
237 |
-
|
238 |
-
ii. a copyright notice;
|
239 |
-
|
240 |
-
iii. a notice that refers to this Public License;
|
241 |
-
|
242 |
-
iv. a notice that refers to the disclaimer of
|
243 |
-
warranties;
|
244 |
-
|
245 |
-
v. a URI or hyperlink to the Licensed Material to the
|
246 |
-
extent reasonably practicable;
|
247 |
-
|
248 |
-
b. indicate if You modified the Licensed Material and
|
249 |
-
retain an indication of any previous modifications; and
|
250 |
-
|
251 |
-
c. indicate the Licensed Material is licensed under this
|
252 |
-
Public License, and include the text of, or the URI or
|
253 |
-
hyperlink to, this Public License.
|
254 |
-
|
255 |
-
2. You may satisfy the conditions in Section 3(a)(1) in any
|
256 |
-
reasonable manner based on the medium, means, and context in
|
257 |
-
which You Share the Licensed Material. For example, it may be
|
258 |
-
reasonable to satisfy the conditions by providing a URI or
|
259 |
-
hyperlink to a resource that includes the required
|
260 |
-
information.
|
261 |
-
|
262 |
-
3. If requested by the Licensor, You must remove any of the
|
263 |
-
information required by Section 3(a)(1)(A) to the extent
|
264 |
-
reasonably practicable.
|
265 |
-
|
266 |
-
4. If You Share Adapted Material You produce, the Adapter's
|
267 |
-
License You apply must not prevent recipients of the Adapted
|
268 |
-
Material from complying with this Public License.
|
269 |
-
|
270 |
-
Section 4 -- Sui Generis Database Rights.
|
271 |
-
|
272 |
-
Where the Licensed Rights include Sui Generis Database Rights that
|
273 |
-
apply to Your use of the Licensed Material:
|
274 |
-
|
275 |
-
a. for the avoidance of doubt, Section 2(a)(1) grants You the right
|
276 |
-
to extract, reuse, reproduce, and Share all or a substantial
|
277 |
-
portion of the contents of the database for NonCommercial purposes
|
278 |
-
only;
|
279 |
-
|
280 |
-
b. if You include all or a substantial portion of the database
|
281 |
-
contents in a database in which You have Sui Generis Database
|
282 |
-
Rights, then the database in which You have Sui Generis Database
|
283 |
-
Rights (but not its individual contents) is Adapted Material; and
|
284 |
-
|
285 |
-
c. You must comply with the conditions in Section 3(a) if You Share
|
286 |
-
all or a substantial portion of the contents of the database.
|
287 |
-
|
288 |
-
For the avoidance of doubt, this Section 4 supplements and does not
|
289 |
-
replace Your obligations under this Public License where the Licensed
|
290 |
-
Rights include other Copyright and Similar Rights.
|
291 |
-
|
292 |
-
Section 5 -- Disclaimer of Warranties and Limitation of Liability.
|
293 |
-
|
294 |
-
a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
|
295 |
-
EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
|
296 |
-
AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
|
297 |
-
ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
|
298 |
-
IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
|
299 |
-
WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
|
300 |
-
PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
|
301 |
-
ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
|
302 |
-
KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
|
303 |
-
ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
|
304 |
-
|
305 |
-
b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
|
306 |
-
TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
|
307 |
-
NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
|
308 |
-
INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
|
309 |
-
COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
|
310 |
-
USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
|
311 |
-
ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
|
312 |
-
DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
|
313 |
-
IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
|
314 |
-
|
315 |
-
c. The disclaimer of warranties and limitation of liability provided
|
316 |
-
above shall be interpreted in a manner that, to the extent
|
317 |
-
possible, most closely approximates an absolute disclaimer and
|
318 |
-
waiver of all liability.
|
319 |
-
|
320 |
-
Section 6 -- Term and Termination.
|
321 |
-
|
322 |
-
a. This Public License applies for the term of the Copyright and
|
323 |
-
Similar Rights licensed here. However, if You fail to comply with
|
324 |
-
this Public License, then Your rights under this Public License
|
325 |
-
terminate automatically.
|
326 |
-
|
327 |
-
b. Where Your right to use the Licensed Material has terminated under
|
328 |
-
Section 6(a), it reinstates:
|
329 |
-
|
330 |
-
1. automatically as of the date the violation is cured, provided
|
331 |
-
it is cured within 30 days of Your discovery of the
|
332 |
-
violation; or
|
333 |
-
|
334 |
-
2. upon express reinstatement by the Licensor.
|
335 |
-
|
336 |
-
For the avoidance of doubt, this Section 6(b) does not affect any
|
337 |
-
right the Licensor may have to seek remedies for Your violations
|
338 |
-
of this Public License.
|
339 |
-
|
340 |
-
c. For the avoidance of doubt, the Licensor may also offer the
|
341 |
-
Licensed Material under separate terms or conditions or stop
|
342 |
-
distributing the Licensed Material at any time; however, doing so
|
343 |
-
will not terminate this Public License.
|
344 |
-
|
345 |
-
d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
|
346 |
-
License.
|
347 |
-
|
348 |
-
Section 7 -- Other Terms and Conditions.
|
349 |
-
|
350 |
-
a. The Licensor shall not be bound by any additional or different
|
351 |
-
terms or conditions communicated by You unless expressly agreed.
|
352 |
-
|
353 |
-
b. Any arrangements, understandings, or agreements regarding the
|
354 |
-
Licensed Material not stated herein are separate from and
|
355 |
-
independent of the terms and conditions of this Public License.
|
356 |
-
|
357 |
-
Section 8 -- Interpretation.
|
358 |
-
|
359 |
-
a. For the avoidance of doubt, this Public License does not, and
|
360 |
-
shall not be interpreted to, reduce, limit, restrict, or impose
|
361 |
-
conditions on any use of the Licensed Material that could lawfully
|
362 |
-
be made without permission under this Public License.
|
363 |
-
|
364 |
-
b. To the extent possible, if any provision of this Public License is
|
365 |
-
deemed unenforceable, it shall be automatically reformed to the
|
366 |
-
minimum extent necessary to make it enforceable. If the provision
|
367 |
-
cannot be reformed, it shall be severed from this Public License
|
368 |
-
without affecting the enforceability of the remaining terms and
|
369 |
-
conditions.
|
370 |
-
|
371 |
-
c. No term or condition of this Public License will be waived and no
|
372 |
-
failure to comply consented to unless expressly agreed to by the
|
373 |
-
Licensor.
|
374 |
-
|
375 |
-
d. Nothing in this Public License constitutes or may be interpreted
|
376 |
-
as a limitation upon, or waiver of, any privileges and immunities
|
377 |
-
that apply to the Licensor or You, including from the legal
|
378 |
-
processes of any jurisdiction or authority.
|
379 |
-
|
380 |
-
=======================================================================
|
381 |
-
|
382 |
-
Creative Commons is not a party to its public
|
383 |
-
licenses. Notwithstanding, Creative Commons may elect to apply one of
|
384 |
-
its public licenses to material it publishes and in those instances
|
385 |
-
will be considered the “Licensor.” The text of the Creative Commons
|
386 |
-
public licenses is dedicated to the public domain under the CC0 Public
|
387 |
-
Domain Dedication. Except for the limited purpose of indicating that
|
388 |
-
material is shared under a Creative Commons public license or as
|
389 |
-
otherwise permitted by the Creative Commons policies published at
|
390 |
-
creativecommons.org/policies, Creative Commons does not authorize the
|
391 |
-
use of the trademark "Creative Commons" or any other trademark or logo
|
392 |
-
of Creative Commons without its prior written consent including,
|
393 |
-
without limitation, in connection with any unauthorized modifications
|
394 |
-
to any of its public licenses or any other arrangements,
|
395 |
-
understandings, or agreements concerning use of licensed material. For
|
396 |
-
the avoidance of doubt, this paragraph does not form part of the
|
397 |
-
public licenses.
|
398 |
-
|
399 |
-
Creative Commons may be contacted at creativecommons.org.
|
|
|
1 |
+
# Attribution-NonCommercial-NoDerivatives 4.0 International
|
2 |
+
|
3 |
+
> *Creative Commons Corporation (“Creative Commons”) is not a law firm and does not provide legal services or legal advice. Distribution of Creative Commons public licenses does not create a lawyer-client or other relationship. Creative Commons makes its licenses and related information available on an “as-is” basis. Creative Commons gives no warranties regarding its licenses, any material licensed under their terms and conditions, or any related information. Creative Commons disclaims all liability for damages resulting from their use to the fullest extent possible.*
|
4 |
+
>
|
5 |
+
> ### Using Creative Commons Public Licenses
|
6 |
+
>
|
7 |
+
> Creative Commons public licenses provide a standard set of terms and conditions that creators and other rights holders may use to share original works of authorship and other material subject to copyright and certain other rights specified in the public license below. The following considerations are for informational purposes only, are not exhaustive, and do not form part of our licenses.
|
8 |
+
>
|
9 |
+
> * __Considerations for licensors:__ Our public licenses are intended for use by those authorized to give the public permission to use material in ways otherwise restricted by copyright and certain other rights. Our licenses are irrevocable. Licensors should read and understand the terms and conditions of the license they choose before applying it. Licensors should also secure all rights necessary before applying our licenses so that the public can reuse the material as expected. Licensors should clearly mark any material not subject to the license. This includes other CC-licensed material, or material used under an exception or limitation to copyright. [More considerations for licensors](http://wiki.creativecommons.org/Considerations_for_licensors_and_licensees#Considerations_for_licensors).
|
10 |
+
>
|
11 |
+
> * __Considerations for the public:__ By using one of our public licenses, a licensor grants the public permission to use the licensed material under specified terms and conditions. If the licensor’s permission is not necessary for any reason–for example, because of any applicable exception or limitation to copyright–then that use is not regulated by the license. Our licenses grant only permissions under copyright and certain other rights that a licensor has authority to grant. Use of the licensed material may still be restricted for other reasons, including because others have copyright or other rights in the material. A licensor may make special requests, such as asking that all changes be marked or described. Although not required by our licenses, you are encouraged to respect those requests where reasonable. [More considerations for the public](http://wiki.creativecommons.org/Considerations_for_licensors_and_licensees#Considerations_for_licensees).
|
12 |
+
|
13 |
+
## Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License
|
14 |
+
|
15 |
+
By exercising the Licensed Rights (defined below), You accept and agree to be bound by the terms and conditions of this Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License ("Public License"). To the extent this Public License may be interpreted as a contract, You are granted the Licensed Rights in consideration of Your acceptance of these terms and conditions, and the Licensor grants You such rights in consideration of benefits the Licensor receives from making the Licensed Material available under these terms and conditions.
|
16 |
+
|
17 |
+
### Section 1 – Definitions.
|
18 |
+
|
19 |
+
a. __Adapted Material__ means material subject to Copyright and Similar Rights that is derived from or based upon the Licensed Material and in which the Licensed Material is translated, altered, arranged, transformed, or otherwise modified in a manner requiring permission under the Copyright and Similar Rights held by the Licensor. For purposes of this Public License, where the Licensed Material is a musical work, performance, or sound recording, Adapted Material is always produced where the Licensed Material is synched in timed relation with a moving image.
|
20 |
+
|
21 |
+
b. __Copyright and Similar Rights__ means copyright and/or similar rights closely related to copyright including, without limitation, performance, broadcast, sound recording, and Sui Generis Database Rights, without regard to how the rights are labeled or categorized. For purposes of this Public License, the rights specified in Section 2(b)(1)-(2) are not Copyright and Similar Rights.
|
22 |
+
|
23 |
+
e. __Effective Technological Measures__ means those measures that, in the absence of proper authority, may not be circumvented under laws fulfilling obligations under Article 11 of the WIPO Copyright Treaty adopted on December 20, 1996, and/or similar international agreements.
|
24 |
+
|
25 |
+
f. __Exceptions and Limitations__ means fair use, fair dealing, and/or any other exception or limitation to Copyright and Similar Rights that applies to Your use of the Licensed Material.
|
26 |
+
|
27 |
+
h. __Licensed Material__ means the artistic or literary work, database, or other material to which the Licensor applied this Public License.
|
28 |
+
|
29 |
+
i. __Licensed Rights__ means the rights granted to You subject to the terms and conditions of this Public License, which are limited to all Copyright and Similar Rights that apply to Your use of the Licensed Material and that the Licensor has authority to license.
|
30 |
+
|
31 |
+
h. __Licensor__ means the individual(s) or entity(ies) granting rights under this Public License.
|
32 |
+
|
33 |
+
i. __NonCommercial__ means not primarily intended for or directed towards commercial advantage or monetary compensation. For purposes of this Public License, the exchange of the Licensed Material for other material subject to Copyright and Similar Rights by digital file-sharing or similar means is NonCommercial provided there is no payment of monetary compensation in connection with the exchange.
|
34 |
+
|
35 |
+
j. __Share__ means to provide material to the public by any means or process that requires permission under the Licensed Rights, such as reproduction, public display, public performance, distribution, dissemination, communication, or importation, and to make material available to the public including in ways that members of the public may access the material from a place and at a time individually chosen by them.
|
36 |
+
|
37 |
+
k. __Sui Generis Database Rights__ means rights other than copyright resulting from Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases, as amended and/or succeeded, as well as other essentially equivalent rights anywhere in the world.
|
38 |
+
|
39 |
+
l. __You__ means the individual or entity exercising the Licensed Rights under this Public License. Your has a corresponding meaning.
|
40 |
+
|
41 |
+
### Section 2 – Scope.
|
42 |
+
|
43 |
+
a. ___License grant.___
|
44 |
+
|
45 |
+
1. Subject to the terms and conditions of this Public License, the Licensor hereby grants You a worldwide, royalty-free, non-sublicensable, non-exclusive, irrevocable license to exercise the Licensed Rights in the Licensed Material to:
|
46 |
+
|
47 |
+
A. reproduce and Share the Licensed Material, in whole or in part, for NonCommercial purposes only; and
|
48 |
+
|
49 |
+
B. produce and reproduce, but not Share, Adapted Material for NonCommercial purposes only.
|
50 |
+
|
51 |
+
2. __Exceptions and Limitations.__ For the avoidance of doubt, where Exceptions and Limitations apply to Your use, this Public License does not apply, and You do not need to comply with its terms and conditions.
|
52 |
+
|
53 |
+
3. __Term.__ The term of this Public License is specified in Section 6(a).
|
54 |
+
|
55 |
+
4. __Media and formats; technical modifications allowed.__ The Licensor authorizes You to exercise the Licensed Rights in all media and formats whether now known or hereafter created, and to make technical modifications necessary to do so. The Licensor waives and/or agrees not to assert any right or authority to forbid You from making technical modifications necessary to exercise the Licensed Rights, including technical modifications necessary to circumvent Effective Technological Measures. For purposes of this Public License, simply making modifications authorized by this Section 2(a)(4) never produces Adapted Material.
|
56 |
+
|
57 |
+
5. __Downstream recipients.__
|
58 |
+
|
59 |
+
A. __Offer from the Licensor – Licensed Material.__ Every recipient of the Licensed Material automatically receives an offer from the Licensor to exercise the Licensed Rights under the terms and conditions of this Public License.
|
60 |
+
|
61 |
+
B. __No downstream restrictions.__ You may not offer or impose any additional or different terms or conditions on, or apply any Effective Technological Measures to, the Licensed Material if doing so restricts exercise of the Licensed Rights by any recipient of the Licensed Material.
|
62 |
+
|
63 |
+
6. __No endorsement.__ Nothing in this Public License constitutes or may be construed as permission to assert or imply that You are, or that Your use of the Licensed Material is, connected with, or sponsored, endorsed, or granted official status by, the Licensor or others designated to receive attribution as provided in Section 3(a)(1)(A)(i).
|
64 |
+
|
65 |
+
b. ___Other rights.___
|
66 |
+
|
67 |
+
1. Moral rights, such as the right of integrity, are not licensed under this Public License, nor are publicity, privacy, and/or other similar personality rights; however, to the extent possible, the Licensor waives and/or agrees not to assert any such rights held by the Licensor to the limited extent necessary to allow You to exercise the Licensed Rights, but not otherwise.
|
68 |
+
|
69 |
+
2. Patent and trademark rights are not licensed under this Public License.
|
70 |
+
|
71 |
+
3. To the extent possible, the Licensor waives any right to collect royalties from You for the exercise of the Licensed Rights, whether directly or through a collecting society under any voluntary or waivable statutory or compulsory licensing scheme. In all other cases the Licensor expressly reserves any right to collect such royalties, including when the Licensed Material is used other than for NonCommercial purposes.
|
72 |
+
|
73 |
+
### Section 3 – License Conditions.
|
74 |
+
|
75 |
+
Your exercise of the Licensed Rights is expressly made subject to the following conditions.
|
76 |
+
|
77 |
+
a. ___Attribution.___
|
78 |
+
|
79 |
+
1. If You Share the Licensed Material, You must:
|
80 |
+
|
81 |
+
A. retain the following if it is supplied by the Licensor with the Licensed Material:
|
82 |
+
|
83 |
+
i. identification of the creator(s) of the Licensed Material and any others designated to receive attribution, in any reasonable manner requested by the Licensor (including by pseudonym if designated);
|
84 |
+
|
85 |
+
ii. a copyright notice;
|
86 |
+
|
87 |
+
iii. a notice that refers to this Public License;
|
88 |
+
|
89 |
+
iv. a notice that refers to the disclaimer of warranties;
|
90 |
+
|
91 |
+
v. a URI or hyperlink to the Licensed Material to the extent reasonably practicable;
|
92 |
+
|
93 |
+
B. indicate if You modified the Licensed Material and retain an indication of any previous modifications; and
|
94 |
+
|
95 |
+
C. indicate the Licensed Material is licensed under this Public License, and include the text of, or the URI or hyperlink to, this Public License.
|
96 |
+
|
97 |
+
For the avoidance of doubt, You do not have permission under this Public License to Share Adapted Material.
|
98 |
+
|
99 |
+
2. You may satisfy the conditions in Section 3(a)(1) in any reasonable manner based on the medium, means, and context in which You Share the Licensed Material. For example, it may be reasonable to satisfy the conditions by providing a URI or hyperlink to a resource that includes the required information.
|
100 |
+
|
101 |
+
3. If requested by the Licensor, You must remove any of the information required by Section 3(a)(1)(A) to the extent reasonably practicable.
|
102 |
+
|
103 |
+
### Section 4 – Sui Generis Database Rights.
|
104 |
+
|
105 |
+
Where the Licensed Rights include Sui Generis Database Rights that apply to Your use of the Licensed Material:
|
106 |
+
|
107 |
+
a. for the avoidance of doubt, Section 2(a)(1) grants You the right to extract, reuse, reproduce, and Share all or a substantial portion of the contents of the database for NonCommercial purposes only and provided You do not Share Adapted Material;
|
108 |
+
|
109 |
+
b. if You include all or a substantial portion of the database contents in a database in which You have Sui Generis Database Rights, then the database in which You have Sui Generis Database Rights (but not its individual contents) is Adapted Material; and
|
110 |
+
|
111 |
+
c. You must comply with the conditions in Section 3(a) if You Share all or a substantial portion of the contents of the database.
|
112 |
+
|
113 |
+
For the avoidance of doubt, this Section 4 supplements and does not replace Your obligations under this Public License where the Licensed Rights include other Copyright and Similar Rights.
|
114 |
+
|
115 |
+
### Section 5 – Disclaimer of Warranties and Limitation of Liability.
|
116 |
+
|
117 |
+
a. __Unless otherwise separately undertaken by the Licensor, to the extent possible, the Licensor offers the Licensed Material as-is and as-available, and makes no representations or warranties of any kind concerning the Licensed Material, whether express, implied, statutory, or other. This includes, without limitation, warranties of title, merchantability, fitness for a particular purpose, non-infringement, absence of latent or other defects, accuracy, or the presence or absence of errors, whether or not known or discoverable. Where disclaimers of warranties are not allowed in full or in part, this disclaimer may not apply to You.__
|
118 |
+
|
119 |
+
b. __To the extent possible, in no event will the Licensor be liable to You on any legal theory (including, without limitation, negligence) or otherwise for any direct, special, indirect, incidental, consequential, punitive, exemplary, or other losses, costs, expenses, or damages arising out of this Public License or use of the Licensed Material, even if the Licensor has been advised of the possibility of such losses, costs, expenses, or damages. Where a limitation of liability is not allowed in full or in part, this limitation may not apply to You.__
|
120 |
+
|
121 |
+
c. The disclaimer of warranties and limitation of liability provided above shall be interpreted in a manner that, to the extent possible, most closely approximates an absolute disclaimer and waiver of all liability.
|
122 |
+
|
123 |
+
### Section 6 – Term and Termination.
|
124 |
+
|
125 |
+
a. This Public License applies for the term of the Copyright and Similar Rights licensed here. However, if You fail to comply with this Public License, then Your rights under this Public License terminate automatically.
|
126 |
+
|
127 |
+
b. Where Your right to use the Licensed Material has terminated under Section 6(a), it reinstates:
|
128 |
+
|
129 |
+
1. automatically as of the date the violation is cured, provided it is cured within 30 days of Your discovery of the violation; or
|
130 |
+
|
131 |
+
2. upon express reinstatement by the Licensor.
|
132 |
+
|
133 |
+
For the avoidance of doubt, this Section 6(b) does not affect any right the Licensor may have to seek remedies for Your violations of this Public License.
|
134 |
+
|
135 |
+
c. For the avoidance of doubt, the Licensor may also offer the Licensed Material under separate terms or conditions or stop distributing the Licensed Material at any time; however, doing so will not terminate this Public License.
|
136 |
+
|
137 |
+
d. Sections 1, 5, 6, 7, and 8 survive termination of this Public License.
|
138 |
+
|
139 |
+
### Section 7 – Other Terms and Conditions.
|
140 |
+
|
141 |
+
a. The Licensor shall not be bound by any additional or different terms or conditions communicated by You unless expressly agreed.
|
142 |
+
|
143 |
+
b. Any arrangements, understandings, or agreements regarding the Licensed Material not stated herein are separate from and independent of the terms and conditions of this Public License.
|
144 |
+
|
145 |
+
### Section 8 – Interpretation.
|
146 |
+
|
147 |
+
a. For the avoidance of doubt, this Public License does not, and shall not be interpreted to, reduce, limit, restrict, or impose conditions on any use of the Licensed Material that could lawfully be made without permission under this Public License.
|
148 |
+
|
149 |
+
b. To the extent possible, if any provision of this Public License is deemed unenforceable, it shall be automatically reformed to the minimum extent necessary to make it enforceable. If the provision cannot be reformed, it shall be severed from this Public License without affecting the enforceability of the remaining terms and conditions.
|
150 |
+
|
151 |
+
c. No term or condition of this Public License will be waived and no failure to comply consented to unless expressly agreed to by the Licensor.
|
152 |
+
|
153 |
+
d. Nothing in this Public License constitutes or may be interpreted as a limitation upon, or waiver of, any privileges and immunities that apply to the Licensor or You, including from the legal processes of any jurisdiction or authority.
|
154 |
+
|
155 |
+
> Creative Commons is not a party to its public licenses. Notwithstanding, Creative Commons may elect to apply one of its public licenses to material it publishes and in those instances will be considered the “Licensor.” Except for the limited purpose of indicating that material is shared under a Creative Commons public license or as otherwise permitted by the Creative Commons policies published at [creativecommons.org/policies](http://creativecommons.org/policies), Creative Commons does not authorize the use of the trademark “Creative Commons” or any other trademark or logo of Creative Commons without its prior written consent including, without limitation, in connection with any unauthorized modifications to any of its public licenses or any other arrangements, understandings, or agreements concerning use of licensed material. For the avoidance of doubt, this paragraph does not form part of the public licenses.
|
156 |
+
>
|
157 |
+
> Creative Commons may be contacted at [creativecommons.org](http://creativecommons.org).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
MANIFEST.in
CHANGED
@@ -6,10 +6,3 @@ include *.ini
|
|
6 |
include requirements.txt
|
7 |
include audiocraft/py.typed
|
8 |
include assets/*.mp3
|
9 |
-
include datasets/*.mp3
|
10 |
-
recursive-include config *.yaml
|
11 |
-
recursive-include demos *.py
|
12 |
-
recursive-include demos *.ipynb
|
13 |
-
recursive-include scripts *.py
|
14 |
-
recursive-include model_cards *.md
|
15 |
-
recursive-include docs *.md
|
|
|
6 |
include requirements.txt
|
7 |
include audiocraft/py.typed
|
8 |
include assets/*.mp3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
model_cards/MUSICGEN_MODEL_CARD.md → MODEL_CARD.md
RENAMED
@@ -12,11 +12,11 @@
|
|
12 |
|
13 |
**Paper or resources for more information:** More information can be found in the paper [Simple and Controllable Music Generation][arxiv].
|
14 |
|
15 |
-
**Citation details
|
16 |
|
17 |
-
**License
|
18 |
|
19 |
-
**Where to send questions or comments about the model:** Questions and comments about MusicGen can be sent via the [
|
20 |
|
21 |
## Intended use
|
22 |
**Primary intended use:** The primary use of MusicGen is research on AI-based music generation, including:
|
@@ -26,7 +26,7 @@
|
|
26 |
|
27 |
**Primary intended users:** The primary intended users of the model are researchers in audio, machine learning and artificial intelligence, as well as amateur seeking to better understand those models.
|
28 |
|
29 |
-
**Out-of-scope use cases
|
30 |
|
31 |
## Metrics
|
32 |
|
@@ -54,24 +54,15 @@ The model was evaluated on the [MusicCaps benchmark](https://www.kaggle.com/data
|
|
54 |
|
55 |
The model was trained on licensed data using the following sources: the [Meta Music Initiative Sound Collection](https://www.fb.com/sound), [Shutterstock music collection](https://www.shutterstock.com/music) and the [Pond5 music collection](https://www.pond5.com/). See the paper for more details about the training set and corresponding preprocessing.
|
56 |
|
57 |
-
##
|
58 |
|
59 |
-
|
60 |
-
|
61 |
-
| Model | Frechet Audio Distance | KLD | Text Consistency | Chroma Cosine Similarity |
|
62 |
-
|---|---|---|---|---|
|
63 |
-
| facebook/musicgen-small | 4.88 | 1.42 | 0.27 | - |
|
64 |
-
| facebook/musicgen-medium | 5.14 | 1.38 | 0.28 | - |
|
65 |
-
| facebook/musicgen-large | 5.48 | 1.37 | 0.28 | - |
|
66 |
-
| facebook/musicgen-melody | 4.93 | 1.41 | 0.27 | 0.44 |
|
67 |
-
|
68 |
-
More information can be found in the paper [Simple and Controllable Music Generation][arxiv], in the Results section.
|
69 |
|
70 |
## Limitations and biases
|
71 |
|
72 |
**Data:** The data sources used to train the model are created by music professionals and covered by legal agreements with the right holders. The model is trained on 20K hours of data, we believe that scaling the model on larger datasets can further improve the performance of the model.
|
73 |
|
74 |
-
**Mitigations:** Vocals have been removed from the data source using corresponding tags, and then using a state-of-the-art music source separation method, namely using the open source [Hybrid Transformer for Music Source Separation](https://github.com/facebookresearch/demucs) (HT-Demucs).
|
75 |
|
76 |
**Limitations:**
|
77 |
|
@@ -87,19 +78,4 @@ More information can be found in the paper [Simple and Controllable Music Genera
|
|
87 |
|
88 |
**Use cases:** Users must be aware of the biases, limitations and risks of the model. MusicGen is a model developed for artificial intelligence research on controllable music generation. As such, it should not be used for downstream applications without further investigation and mitigation of risks.
|
89 |
|
90 |
-
## Update: stereo models and large melody.
|
91 |
-
|
92 |
-
We further release a set of stereophonic capable models. Those were fine tuned for 200k updates starting
|
93 |
-
from the mono models. The training data is otherwise identical and capabilities and limitations are shared with the base modes. The stereo models work by getting 2 streams of tokens from the EnCodec model, and interleaving those using
|
94 |
-
the delay pattern. We also release a mono large model with melody conditioning capabilities. The list of new models
|
95 |
-
is as follow:
|
96 |
-
|
97 |
-
- facebook/musicgen-stereo-small
|
98 |
-
- facebook/musicgen-stereo-medium
|
99 |
-
- facebook/musicgen-stereo-large
|
100 |
-
- facebook/musicgen-stereo-melody
|
101 |
-
- facebook/musicgen-melody-large
|
102 |
-
- facebook/musicgen-stereo-melody-large
|
103 |
-
|
104 |
-
|
105 |
[arxiv]: https://arxiv.org/abs/2306.05284
|
|
|
12 |
|
13 |
**Paper or resources for more information:** More information can be found in the paper [Simple and Controllable Music Generation][arxiv].
|
14 |
|
15 |
+
**Citation details** See [our paper][arxiv]
|
16 |
|
17 |
+
**License** Code is released under MIT, model weights are released under CC-BY-NC 4.0.
|
18 |
|
19 |
+
**Where to send questions or comments about the model:** Questions and comments about MusicGen can be sent via the [Github repository](https://github.com/facebookresearch/audiocraft) of the project, or by opening an issue.
|
20 |
|
21 |
## Intended use
|
22 |
**Primary intended use:** The primary use of MusicGen is research on AI-based music generation, including:
|
|
|
26 |
|
27 |
**Primary intended users:** The primary intended users of the model are researchers in audio, machine learning and artificial intelligence, as well as amateur seeking to better understand those models.
|
28 |
|
29 |
+
**Out-of-scope use cases** The model should not be used on downstream applications without further risk evaluation and mitigation. The model should not be used to intentionally create or disseminate music pieces that create hostile or alienating environments for people. This includes generating music that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes.
|
30 |
|
31 |
## Metrics
|
32 |
|
|
|
54 |
|
55 |
The model was trained on licensed data using the following sources: the [Meta Music Initiative Sound Collection](https://www.fb.com/sound), [Shutterstock music collection](https://www.shutterstock.com/music) and the [Pond5 music collection](https://www.pond5.com/). See the paper for more details about the training set and corresponding preprocessing.
|
56 |
|
57 |
+
## Quantitative analysis
|
58 |
|
59 |
+
More information can be found in the paper [Simple and Controllable Music Generation][arxiv], in the Experimental Setup section.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
60 |
|
61 |
## Limitations and biases
|
62 |
|
63 |
**Data:** The data sources used to train the model are created by music professionals and covered by legal agreements with the right holders. The model is trained on 20K hours of data, we believe that scaling the model on larger datasets can further improve the performance of the model.
|
64 |
|
65 |
+
**Mitigations:** Vocals have been removed from the data source using corresponding tags, and then using using a state-of-the-art music source separation method, namely using the open source [Hybrid Transformer for Music Source Separation](https://github.com/facebookresearch/demucs) (HT-Demucs).
|
66 |
|
67 |
**Limitations:**
|
68 |
|
|
|
78 |
|
79 |
**Use cases:** Users must be aware of the biases, limitations and risks of the model. MusicGen is a model developed for artificial intelligence research on controllable music generation. As such, it should not be used for downstream applications without further investigation and mitigation of risks.
|
80 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
[arxiv]: https://arxiv.org/abs/2306.05284
|
Makefile
CHANGED
@@ -1,15 +1,3 @@
|
|
1 |
-
INTEG=AUDIOCRAFT_DORA_DIR="/tmp/magma_$(USER)" python3 -m dora -v run --clear device=cpu dataset.num_workers=0 optim.epochs=1 \
|
2 |
-
dataset.train.num_samples=10 dataset.valid.num_samples=10 \
|
3 |
-
dataset.evaluate.num_samples=10 dataset.generate.num_samples=2 sample_rate=16000 \
|
4 |
-
logging.level=DEBUG
|
5 |
-
INTEG_COMPRESSION = $(INTEG) solver=compression/debug rvq.n_q=2 rvq.bins=48 checkpoint.save_last=true # SIG is 5091833e
|
6 |
-
INTEG_MUSICGEN = $(INTEG) solver=musicgen/debug dset=audio/example compression_model_checkpoint=//sig/5091833e \
|
7 |
-
transformer_lm.n_q=2 transformer_lm.card=48 transformer_lm.dim=16 checkpoint.save_last=false # Using compression model from 5091833e
|
8 |
-
INTEG_AUDIOGEN = $(INTEG) solver=audiogen/debug dset=audio/example compression_model_checkpoint=//sig/5091833e \
|
9 |
-
transformer_lm.n_q=2 transformer_lm.card=48 transformer_lm.dim=16 checkpoint.save_last=false # Using compression model from 5091833e
|
10 |
-
INTEG_MBD = $(INTEG) solver=diffusion/debug dset=audio/example \
|
11 |
-
checkpoint.save_last=false # Using compression model from 616d7b3c
|
12 |
-
|
13 |
default: linter tests
|
14 |
|
15 |
install:
|
@@ -22,19 +10,12 @@ linter:
|
|
22 |
|
23 |
tests:
|
24 |
coverage run -m pytest tests
|
25 |
-
coverage report
|
26 |
-
|
27 |
-
tests_integ:
|
28 |
-
$(INTEG_COMPRESSION)
|
29 |
-
$(INTEG_MBD)
|
30 |
-
$(INTEG_MUSICGEN)
|
31 |
-
$(INTEG_AUDIOGEN)
|
32 |
-
|
33 |
|
34 |
-
|
35 |
-
pdoc3 --html -o
|
36 |
|
37 |
dist:
|
38 |
python setup.py sdist
|
39 |
|
40 |
-
.PHONY: linter tests
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
default: linter tests
|
2 |
|
3 |
install:
|
|
|
10 |
|
11 |
tests:
|
12 |
coverage run -m pytest tests
|
13 |
+
coverage report --include 'audiocraft/*'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
+
docs:
|
16 |
+
pdoc3 --html -o docs -f audiocraft
|
17 |
|
18 |
dist:
|
19 |
python setup.py sdist
|
20 |
|
21 |
+
.PHONY: linter tests docs dist
|
README.md
CHANGED
@@ -5,27 +5,42 @@ tags:
|
|
5 |
- "music generation"
|
6 |
- "language models"
|
7 |
- "LLMs"
|
8 |
-
app_file: "
|
9 |
emoji: 🎵
|
10 |
-
colorFrom:
|
11 |
colorTo: blue
|
12 |
sdk: gradio
|
13 |
sdk_version: 3.34.0
|
14 |
pinned: true
|
15 |
license: "cc-by-nc-4.0"
|
16 |
-
disable_embedding: true
|
17 |
---
|
18 |
-
#
|
19 |
![docs badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_docs/badge.svg)
|
20 |
![linter badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_linter/badge.svg)
|
21 |
![tests badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_tests/badge.svg)
|
22 |
|
23 |
-
|
24 |
-
for two state-of-the-art AI generative models producing high-quality audio: AudioGen and MusicGen.
|
25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
|
27 |
## Installation
|
28 |
-
|
29 |
|
30 |
```shell
|
31 |
# Best to make sure you have torch installed first, in particular before installing xformers.
|
@@ -34,68 +49,87 @@ pip install 'torch>=2.0'
|
|
34 |
# Then proceed to one of the following
|
35 |
pip install -U audiocraft # stable release
|
36 |
pip install -U git+https://git@github.com/facebookresearch/audiocraft#egg=audiocraft # bleeding edge
|
37 |
-
pip install -e . # or if you cloned the repo locally
|
38 |
```
|
39 |
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
|
|
46 |
|
47 |
-
##
|
48 |
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
|
55 |
-
|
|
|
|
|
56 |
|
57 |
-
|
58 |
-
|
59 |
-
|
|
|
|
|
60 |
|
61 |
-
|
62 |
-
that provides pointers to configuration, example grids and model/task-specific information and FAQ.
|
63 |
|
|
|
|
|
|
|
|
|
64 |
|
65 |
-
|
|
|
|
|
|
|
|
|
66 |
|
67 |
-
|
|
|
|
|
68 |
|
|
|
|
|
|
|
|
|
69 |
|
70 |
-
## FAQ
|
71 |
|
72 |
-
|
73 |
|
74 |
-
|
75 |
|
76 |
-
|
77 |
|
78 |
-
|
79 |
-
In order to change the cache location of the other Hugging Face models, please check out the [Hugging Face Transformers documentation for the cache setup](https://huggingface.co/docs/transformers/installation#cache-setup).
|
80 |
-
Finally, if you use a model that relies on Demucs (e.g. `musicgen-melody`) and want to change the download location for Demucs, refer to the [Torch Hub documentation](https://pytorch.org/docs/stable/hub.html#where-are-my-downloaded-models-saved).
|
81 |
|
|
|
82 |
|
83 |
-
## License
|
84 |
-
* The code in this repository is released under the MIT license as found in the [LICENSE file](LICENSE).
|
85 |
-
* The models weights in this repository are released under the CC-BY-NC 4.0 license as found in the [LICENSE_weights file](LICENSE_weights).
|
86 |
|
|
|
|
|
|
|
87 |
|
88 |
-
## Citation
|
89 |
|
90 |
-
|
91 |
```
|
92 |
@article{copet2023simple,
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
}
|
98 |
```
|
99 |
|
100 |
-
|
101 |
-
[
|
|
|
|
|
|
|
|
|
|
5 |
- "music generation"
|
6 |
- "language models"
|
7 |
- "LLMs"
|
8 |
+
app_file: "app_batched.py"
|
9 |
emoji: 🎵
|
10 |
+
colorFrom: white
|
11 |
colorTo: blue
|
12 |
sdk: gradio
|
13 |
sdk_version: 3.34.0
|
14 |
pinned: true
|
15 |
license: "cc-by-nc-4.0"
|
|
|
16 |
---
|
17 |
+
# Audiocraft
|
18 |
![docs badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_docs/badge.svg)
|
19 |
![linter badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_linter/badge.svg)
|
20 |
![tests badge](https://github.com/facebookresearch/audiocraft/workflows/audiocraft_tests/badge.svg)
|
21 |
|
22 |
+
Audiocraft is a PyTorch library for deep learning research on audio generation. At the moment, it contains the code for MusicGen, a state-of-the-art controllable text-to-music model.
|
|
|
23 |
|
24 |
+
## MusicGen
|
25 |
+
|
26 |
+
Audiocraft provides the code and models for MusicGen, [a simple and controllable model for music generation][arxiv]. MusicGen is a single stage auto-regressive
|
27 |
+
Transformer model trained over a 32kHz <a href="https://github.com/facebookresearch/encodec">EnCodec tokenizer</a> with 4 codebooks sampled at 50 Hz. Unlike existing methods like [MusicLM](https://arxiv.org/abs/2301.11325), MusicGen doesn't require a self-supervised semantic representation, and it generates
|
28 |
+
all 4 codebooks in one pass. By introducing a small delay between the codebooks, we show we can predict
|
29 |
+
them in parallel, thus having only 50 auto-regressive steps per second of audio.
|
30 |
+
Check out our [sample page][musicgen_samples] or test the available demo!
|
31 |
+
|
32 |
+
<a target="_blank" href="https://colab.research.google.com/drive/1-Xe9NCdIs2sCUbiSmwHXozK6AAhMm7_i?usp=sharing">
|
33 |
+
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
|
34 |
+
</a>
|
35 |
+
<a target="_blank" href="https://huggingface.co/spaces/facebook/MusicGen">
|
36 |
+
<img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-sm.svg" alt="Open in HugginFace"/>
|
37 |
+
</a>
|
38 |
+
<br>
|
39 |
+
|
40 |
+
We use 20K hours of licensed music to train MusicGen. Specifically, we rely on an internal dataset of 10K high-quality music tracks, and on the ShutterStock and Pond5 music data.
|
41 |
|
42 |
## Installation
|
43 |
+
Audiocraft requires Python 3.9, PyTorch 2.0.0, and a GPU with at least 16 GB of memory (for the medium-sized model). To install Audiocraft, you can run the following:
|
44 |
|
45 |
```shell
|
46 |
# Best to make sure you have torch installed first, in particular before installing xformers.
|
|
|
49 |
# Then proceed to one of the following
|
50 |
pip install -U audiocraft # stable release
|
51 |
pip install -U git+https://git@github.com/facebookresearch/audiocraft#egg=audiocraft # bleeding edge
|
52 |
+
pip install -e . # or if you cloned the repo locally
|
53 |
```
|
54 |
|
55 |
+
## Usage
|
56 |
+
We offer a number of way to interact with MusicGen:
|
57 |
+
1. You can play with MusicGen by running the jupyter notebook at [`demo.ipynb`](./demo.ipynb) locally, or use the provided [colab notebook](https://colab.research.google.com/drive/1fxGqfg96RBUvGxZ1XXN07s3DthrKUl4-?usp=sharing).
|
58 |
+
2. You can use the gradio demo locally by running `python app.py`.
|
59 |
+
3. A demo is also available on the [`facebook/MusicGen` HuggingFace Space](https://huggingface.co/spaces/facebook/MusicGen) (huge thanks to all the HF team for their support).
|
60 |
+
4. Finally, you can run the [Gradio demo with a Colab GPU](https://colab.research.google.com/drive/1-Xe9NCdIs2sCUbiSmwHXozK6AAhMm7_i?usp=sharing),
|
61 |
+
as adapted from [@camenduru Colab](https://github.com/camenduru/MusicGen-colab).
|
62 |
|
63 |
+
## API
|
64 |
|
65 |
+
We provide a simple API and 4 pre-trained models. The pre trained models are:
|
66 |
+
- `small`: 300M model, text to music only - [🤗 Hub](https://huggingface.co/facebook/musicgen-small)
|
67 |
+
- `medium`: 1.5B model, text to music only - [🤗 Hub](https://huggingface.co/facebook/musicgen-medium)
|
68 |
+
- `melody`: 1.5B model, text to music and text+melody to music - [🤗 Hub](https://huggingface.co/facebook/musicgen-melody)
|
69 |
+
- `large`: 3.3B model, text to music only - [🤗 Hub](https://huggingface.co/facebook/musicgen-large)
|
70 |
|
71 |
+
We observe the best trade-off between quality and compute with the `medium` or `melody` model.
|
72 |
+
In order to use MusicGen locally **you must have a GPU**. We recommend 16GB of memory, but smaller
|
73 |
+
GPUs will be able to generate short sequences, or longer sequences with the `small` model.
|
74 |
|
75 |
+
**Note**: Please make sure to have [ffmpeg](https://ffmpeg.org/download.html) installed when using newer version of `torchaudio`.
|
76 |
+
You can install it with:
|
77 |
+
```
|
78 |
+
apt-get install ffmpeg
|
79 |
+
```
|
80 |
|
81 |
+
See after a quick example for using the API.
|
|
|
82 |
|
83 |
+
```python
|
84 |
+
import torchaudio
|
85 |
+
from audiocraft.models import MusicGen
|
86 |
+
from audiocraft.data.audio import audio_write
|
87 |
|
88 |
+
model = MusicGen.get_pretrained('melody')
|
89 |
+
model.set_generation_params(duration=8) # generate 8 seconds.
|
90 |
+
wav = model.generate_unconditional(4) # generates 4 unconditional audio samples
|
91 |
+
descriptions = ['happy rock', 'energetic EDM', 'sad jazz']
|
92 |
+
wav = model.generate(descriptions) # generates 3 samples.
|
93 |
|
94 |
+
melody, sr = torchaudio.load('./assets/bach.mp3')
|
95 |
+
# generates using the melody from the given audio and the provided descriptions.
|
96 |
+
wav = model.generate_with_chroma(descriptions, melody[None].expand(3, -1, -1), sr)
|
97 |
|
98 |
+
for idx, one_wav in enumerate(wav):
|
99 |
+
# Will save under {idx}.wav, with loudness normalization at -14 db LUFS.
|
100 |
+
audio_write(f'{idx}', one_wav.cpu(), model.sample_rate, strategy="loudness", loudness_compressor=True)
|
101 |
+
```
|
102 |
|
|
|
103 |
|
104 |
+
## Model Card
|
105 |
|
106 |
+
See [the model card page](./MODEL_CARD.md).
|
107 |
|
108 |
+
## FAQ
|
109 |
|
110 |
+
#### Will the training code be released?
|
|
|
|
|
111 |
|
112 |
+
Yes. We will soon release the training code for MusicGen and EnCodec.
|
113 |
|
|
|
|
|
|
|
114 |
|
115 |
+
#### I need help on Windows
|
116 |
+
|
117 |
+
@FurkanGozukara made a complete tutorial for [Audiocraft/MusicGen on Windows](https://youtu.be/v-YpvPkhdO4)
|
118 |
|
|
|
119 |
|
120 |
+
## Citation
|
121 |
```
|
122 |
@article{copet2023simple,
|
123 |
+
title={Simple and Controllable Music Generation},
|
124 |
+
author={Jade Copet and Felix Kreuk and Itai Gat and Tal Remez and David Kant and Gabriel Synnaeve and Yossi Adi and Alexandre Défossez},
|
125 |
+
year={2023},
|
126 |
+
journal={arXiv preprint arXiv:2306.05284},
|
127 |
}
|
128 |
```
|
129 |
|
130 |
+
## License
|
131 |
+
* The code in this repository is released under the MIT license as found in the [LICENSE file](LICENSE).
|
132 |
+
* The weights in this repository are released under the CC-BY-NC 4.0 license as found in the [LICENSE_weights file](LICENSE_weights).
|
133 |
+
|
134 |
+
[arxiv]: https://arxiv.org/abs/2306.05284
|
135 |
+
[musicgen_samples]: https://ai.honu.io/papers/musicgen/
|
app.py
ADDED
@@ -0,0 +1,216 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""
|
2 |
+
Copyright (c) Meta Platforms, Inc. and affiliates.
|
3 |
+
All rights reserved.
|
4 |
+
|
5 |
+
This source code is licensed under the license found in the
|
6 |
+
LICENSE file in the root directory of this source tree.
|
7 |
+
"""
|
8 |
+
|
9 |
+
from tempfile import NamedTemporaryFile
|
10 |
+
import argparse
|
11 |
+
import torch
|
12 |
+
import gradio as gr
|
13 |
+
import os
|
14 |
+
from audiocraft.models import MusicGen
|
15 |
+
from audiocraft.data.audio import audio_write
|
16 |
+
|
17 |
+
MODEL = None
|
18 |
+
IS_SHARED_SPACE = "musicgen/MusicGen" in os.environ.get('SPACE_ID', '')
|
19 |
+
|
20 |
+
|
21 |
+
def load_model(version):
|
22 |
+
print("Loading model", version)
|
23 |
+
return MusicGen.get_pretrained(version)
|
24 |
+
|
25 |
+
|
26 |
+
def predict(model, text, melody, duration, topk, topp, temperature, cfg_coef):
|
27 |
+
global MODEL
|
28 |
+
topk = int(topk)
|
29 |
+
if MODEL is None or MODEL.name != model:
|
30 |
+
MODEL = load_model(model)
|
31 |
+
|
32 |
+
if duration > MODEL.lm.cfg.dataset.segment_duration:
|
33 |
+
raise gr.Error("MusicGen currently supports durations of up to 30 seconds!")
|
34 |
+
MODEL.set_generation_params(
|
35 |
+
use_sampling=True,
|
36 |
+
top_k=topk,
|
37 |
+
top_p=topp,
|
38 |
+
temperature=temperature,
|
39 |
+
cfg_coef=cfg_coef,
|
40 |
+
duration=duration,
|
41 |
+
)
|
42 |
+
|
43 |
+
if melody:
|
44 |
+
sr, melody = melody[0], torch.from_numpy(melody[1]).to(MODEL.device).float().t().unsqueeze(0)
|
45 |
+
print(melody.shape)
|
46 |
+
if melody.dim() == 2:
|
47 |
+
melody = melody[None]
|
48 |
+
melody = melody[..., :int(sr * MODEL.lm.cfg.dataset.segment_duration)]
|
49 |
+
output = MODEL.generate_with_chroma(
|
50 |
+
descriptions=[text],
|
51 |
+
melody_wavs=melody,
|
52 |
+
melody_sample_rate=sr,
|
53 |
+
progress=False
|
54 |
+
)
|
55 |
+
else:
|
56 |
+
output = MODEL.generate(descriptions=[text], progress=False)
|
57 |
+
|
58 |
+
output = output.detach().cpu().float()[0]
|
59 |
+
with NamedTemporaryFile("wb", suffix=".wav", delete=False) as file:
|
60 |
+
audio_write(
|
61 |
+
file.name, output, MODEL.sample_rate, strategy="loudness",
|
62 |
+
loudness_headroom_db=16, loudness_compressor=True, add_suffix=False)
|
63 |
+
waveform_video = gr.make_waveform(file.name)
|
64 |
+
return waveform_video
|
65 |
+
|
66 |
+
|
67 |
+
def ui(**kwargs):
|
68 |
+
with gr.Blocks() as interface:
|
69 |
+
gr.Markdown(
|
70 |
+
"""
|
71 |
+
# MusicGen
|
72 |
+
This is your private demo for [MusicGen](https://github.com/facebookresearch/audiocraft), a simple and controllable model for music generation
|
73 |
+
presented at: ["Simple and Controllable Music Generation"](https://huggingface.co/papers/2306.05284)
|
74 |
+
"""
|
75 |
+
)
|
76 |
+
if IS_SHARED_SPACE:
|
77 |
+
gr.Markdown("""
|
78 |
+
⚠ This Space doesn't work in this shared UI ⚠
|
79 |
+
|
80 |
+
<a href="https://huggingface.co/spaces/musicgen/MusicGen?duplicate=true" style="display: inline-block;margin-top: .5em;margin-right: .25em;" target="_blank">
|
81 |
+
<img style="margin-bottom: 0em;display: inline;margin-top: -.25em;" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
82 |
+
to use it privately, or use the <a href="https://huggingface.co/spaces/facebook/MusicGen">public demo</a>
|
83 |
+
""")
|
84 |
+
with gr.Row():
|
85 |
+
with gr.Column():
|
86 |
+
with gr.Row():
|
87 |
+
text = gr.Text(label="Input Text", interactive=True)
|
88 |
+
melody = gr.Audio(source="upload", type="numpy", label="Melody Condition (optional)", interactive=True)
|
89 |
+
with gr.Row():
|
90 |
+
submit = gr.Button("Submit")
|
91 |
+
with gr.Row():
|
92 |
+
model = gr.Radio(["melody", "medium", "small", "large"], label="Model", value="melody", interactive=True)
|
93 |
+
with gr.Row():
|
94 |
+
duration = gr.Slider(minimum=1, maximum=30, value=10, label="Duration", interactive=True)
|
95 |
+
with gr.Row():
|
96 |
+
topk = gr.Number(label="Top-k", value=250, interactive=True)
|
97 |
+
topp = gr.Number(label="Top-p", value=0, interactive=True)
|
98 |
+
temperature = gr.Number(label="Temperature", value=1.0, interactive=True)
|
99 |
+
cfg_coef = gr.Number(label="Classifier Free Guidance", value=3.0, interactive=True)
|
100 |
+
with gr.Column():
|
101 |
+
output = gr.Video(label="Generated Music")
|
102 |
+
submit.click(predict, inputs=[model, text, melody, duration, topk, topp, temperature, cfg_coef], outputs=[output])
|
103 |
+
gr.Examples(
|
104 |
+
fn=predict,
|
105 |
+
examples=[
|
106 |
+
[
|
107 |
+
"An 80s driving pop song with heavy drums and synth pads in the background",
|
108 |
+
"./assets/bach.mp3",
|
109 |
+
"melody"
|
110 |
+
],
|
111 |
+
[
|
112 |
+
"A cheerful country song with acoustic guitars",
|
113 |
+
"./assets/bolero_ravel.mp3",
|
114 |
+
"melody"
|
115 |
+
],
|
116 |
+
[
|
117 |
+
"90s rock song with electric guitar and heavy drums",
|
118 |
+
None,
|
119 |
+
"medium"
|
120 |
+
],
|
121 |
+
[
|
122 |
+
"a light and cheerly EDM track, with syncopated drums, aery pads, and strong emotions",
|
123 |
+
"./assets/bach.mp3",
|
124 |
+
"melody"
|
125 |
+
],
|
126 |
+
[
|
127 |
+
"lofi slow bpm electro chill with organic samples",
|
128 |
+
None,
|
129 |
+
"medium",
|
130 |
+
],
|
131 |
+
],
|
132 |
+
inputs=[text, melody, model],
|
133 |
+
outputs=[output]
|
134 |
+
)
|
135 |
+
gr.Markdown(
|
136 |
+
"""
|
137 |
+
### More details
|
138 |
+
|
139 |
+
The model will generate a short music extract based on the description you provided.
|
140 |
+
You can generate up to 30 seconds of audio.
|
141 |
+
|
142 |
+
We present 4 model variations:
|
143 |
+
1. Melody -- a music generation model capable of generating music condition on text and melody inputs. **Note**, you can also use text only.
|
144 |
+
2. Small -- a 300M transformer decoder conditioned on text only.
|
145 |
+
3. Medium -- a 1.5B transformer decoder conditioned on text only.
|
146 |
+
4. Large -- a 3.3B transformer decoder conditioned on text only (might OOM for the longest sequences.)
|
147 |
+
|
148 |
+
When using `melody`, ou can optionaly provide a reference audio from
|
149 |
+
which a broad melody will be extracted. The model will then try to follow both the description and melody provided.
|
150 |
+
|
151 |
+
You can also use your own GPU or a Google Colab by following the instructions on our repo.
|
152 |
+
See [github.com/facebookresearch/audiocraft](https://github.com/facebookresearch/audiocraft)
|
153 |
+
for more details.
|
154 |
+
"""
|
155 |
+
)
|
156 |
+
|
157 |
+
# Show the interface
|
158 |
+
launch_kwargs = {}
|
159 |
+
username = kwargs.get('username')
|
160 |
+
password = kwargs.get('password')
|
161 |
+
server_port = kwargs.get('server_port', 0)
|
162 |
+
inbrowser = kwargs.get('inbrowser', False)
|
163 |
+
share = kwargs.get('share', False)
|
164 |
+
server_name = kwargs.get('listen')
|
165 |
+
|
166 |
+
launch_kwargs['server_name'] = server_name
|
167 |
+
|
168 |
+
if username and password:
|
169 |
+
launch_kwargs['auth'] = (username, password)
|
170 |
+
if server_port > 0:
|
171 |
+
launch_kwargs['server_port'] = server_port
|
172 |
+
if inbrowser:
|
173 |
+
launch_kwargs['inbrowser'] = inbrowser
|
174 |
+
if share:
|
175 |
+
launch_kwargs['share'] = share
|
176 |
+
|
177 |
+
interface.queue().launch(**launch_kwargs, max_threads=1)
|
178 |
+
|
179 |
+
|
180 |
+
if __name__ == "__main__":
|
181 |
+
parser = argparse.ArgumentParser()
|
182 |
+
parser.add_argument(
|
183 |
+
'--listen',
|
184 |
+
type=str,
|
185 |
+
default='0.0.0.0',
|
186 |
+
help='IP to listen on for connections to Gradio',
|
187 |
+
)
|
188 |
+
parser.add_argument(
|
189 |
+
'--username', type=str, default='', help='Username for authentication'
|
190 |
+
)
|
191 |
+
parser.add_argument(
|
192 |
+
'--password', type=str, default='', help='Password for authentication'
|
193 |
+
)
|
194 |
+
parser.add_argument(
|
195 |
+
'--server_port',
|
196 |
+
type=int,
|
197 |
+
default=0,
|
198 |
+
help='Port to run the server listener on',
|
199 |
+
)
|
200 |
+
parser.add_argument(
|
201 |
+
'--inbrowser', action='store_true', help='Open in browser'
|
202 |
+
)
|
203 |
+
parser.add_argument(
|
204 |
+
'--share', action='store_true', help='Share the gradio UI'
|
205 |
+
)
|
206 |
+
|
207 |
+
args = parser.parse_args()
|
208 |
+
|
209 |
+
ui(
|
210 |
+
username=args.username,
|
211 |
+
password=args.password,
|
212 |
+
inbrowser=args.inbrowser,
|
213 |
+
server_port=args.server_port,
|
214 |
+
share=args.share,
|
215 |
+
listen=args.listen
|
216 |
+
)
|
app_batched.py
ADDED
@@ -0,0 +1,222 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""
|
2 |
+
Copyright (c) Meta Platforms, Inc. and affiliates.
|
3 |
+
All rights reserved.
|
4 |
+
|
5 |
+
This source code is licensed under the license found in the
|
6 |
+
LICENSE file in the root directory of this source tree.
|
7 |
+
"""
|
8 |
+
|
9 |
+
import argparse
|
10 |
+
from concurrent.futures import ProcessPoolExecutor
|
11 |
+
import subprocess as sp
|
12 |
+
from tempfile import NamedTemporaryFile
|
13 |
+
import time
|
14 |
+
import warnings
|
15 |
+
import torch
|
16 |
+
import gradio as gr
|
17 |
+
from audiocraft.data.audio_utils import convert_audio
|
18 |
+
from audiocraft.data.audio import audio_write
|
19 |
+
from audiocraft.models import MusicGen
|
20 |
+
|
21 |
+
|
22 |
+
MODEL = None
|
23 |
+
|
24 |
+
_old_call = sp.call
|
25 |
+
|
26 |
+
|
27 |
+
def _call_nostderr(*args, **kwargs):
|
28 |
+
# Avoid ffmpeg vomitting on the logs.
|
29 |
+
kwargs['stderr'] = sp.DEVNULL
|
30 |
+
kwargs['stdout'] = sp.DEVNULL
|
31 |
+
_old_call(*args, **kwargs)
|
32 |
+
|
33 |
+
|
34 |
+
sp.call = _call_nostderr
|
35 |
+
pool = ProcessPoolExecutor(3)
|
36 |
+
pool.__enter__()
|
37 |
+
|
38 |
+
|
39 |
+
def make_waveform(*args, **kwargs):
|
40 |
+
be = time.time()
|
41 |
+
with warnings.catch_warnings():
|
42 |
+
warnings.simplefilter('ignore')
|
43 |
+
out = gr.make_waveform(*args, **kwargs)
|
44 |
+
print("Make a video took", time.time() - be)
|
45 |
+
return out
|
46 |
+
|
47 |
+
|
48 |
+
def load_model():
|
49 |
+
print("Loading model")
|
50 |
+
return MusicGen.get_pretrained("melody")
|
51 |
+
|
52 |
+
|
53 |
+
def predict(texts, melodies):
|
54 |
+
global MODEL
|
55 |
+
if MODEL is None:
|
56 |
+
MODEL = load_model()
|
57 |
+
|
58 |
+
duration = 12
|
59 |
+
max_text_length = 512
|
60 |
+
texts = [text[:max_text_length] for text in texts]
|
61 |
+
MODEL.set_generation_params(duration=duration)
|
62 |
+
|
63 |
+
print("new batch", len(texts), texts, [None if m is None else (m[0], m[1].shape) for m in melodies])
|
64 |
+
be = time.time()
|
65 |
+
processed_melodies = []
|
66 |
+
target_sr = 32000
|
67 |
+
target_ac = 1
|
68 |
+
for melody in melodies:
|
69 |
+
if melody is None:
|
70 |
+
processed_melodies.append(None)
|
71 |
+
else:
|
72 |
+
sr, melody = melody[0], torch.from_numpy(melody[1]).to(MODEL.device).float().t()
|
73 |
+
if melody.dim() == 1:
|
74 |
+
melody = melody[None]
|
75 |
+
melody = melody[..., :int(sr * duration)]
|
76 |
+
melody = convert_audio(melody, sr, target_sr, target_ac)
|
77 |
+
processed_melodies.append(melody)
|
78 |
+
|
79 |
+
outputs = MODEL.generate_with_chroma(
|
80 |
+
descriptions=texts,
|
81 |
+
melody_wavs=processed_melodies,
|
82 |
+
melody_sample_rate=target_sr,
|
83 |
+
progress=False
|
84 |
+
)
|
85 |
+
|
86 |
+
outputs = outputs.detach().cpu().float()
|
87 |
+
out_files = []
|
88 |
+
for output in outputs:
|
89 |
+
with NamedTemporaryFile("wb", suffix=".wav", delete=False) as file:
|
90 |
+
audio_write(
|
91 |
+
file.name, output, MODEL.sample_rate, strategy="loudness",
|
92 |
+
loudness_headroom_db=16, loudness_compressor=True, add_suffix=False)
|
93 |
+
out_files.append(pool.submit(make_waveform, file.name))
|
94 |
+
res = [[out_file.result() for out_file in out_files]]
|
95 |
+
print("batch finished", len(texts), time.time() - be)
|
96 |
+
return res
|
97 |
+
|
98 |
+
|
99 |
+
def ui(**kwargs):
|
100 |
+
with gr.Blocks() as demo:
|
101 |
+
gr.Markdown(
|
102 |
+
"""
|
103 |
+
# MusicGen
|
104 |
+
|
105 |
+
This is the demo for [MusicGen](https://github.com/facebookresearch/audiocraft), a simple and controllable model for music generation
|
106 |
+
presented at: ["Simple and Controllable Music Generation"](https://huggingface.co/papers/2306.05284).
|
107 |
+
<br/>
|
108 |
+
<a href="https://huggingface.co/spaces/musicgen/MusicGen?duplicate=true" style="display: inline-block;margin-top: .5em;margin-right: .25em;" target="_blank">
|
109 |
+
<img style="margin-bottom: 0em;display: inline;margin-top: -.25em;" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
110 |
+
for longer sequences, more control and no queue.</p>
|
111 |
+
"""
|
112 |
+
)
|
113 |
+
with gr.Row():
|
114 |
+
with gr.Column():
|
115 |
+
with gr.Row():
|
116 |
+
text = gr.Text(label="Describe your music", lines=2, interactive=True)
|
117 |
+
melody = gr.Audio(source="upload", type="numpy", label="Condition on a melody (optional)", interactive=True)
|
118 |
+
with gr.Row():
|
119 |
+
submit = gr.Button("Generate")
|
120 |
+
with gr.Column():
|
121 |
+
output = gr.Video(label="Generated Music")
|
122 |
+
submit.click(predict, inputs=[text, melody], outputs=[output], batch=True, max_batch_size=8)
|
123 |
+
gr.Examples(
|
124 |
+
fn=predict,
|
125 |
+
examples=[
|
126 |
+
[
|
127 |
+
"An 80s driving pop song with heavy drums and synth pads in the background",
|
128 |
+
"./assets/bach.mp3",
|
129 |
+
],
|
130 |
+
[
|
131 |
+
"A cheerful country song with acoustic guitars",
|
132 |
+
"./assets/bolero_ravel.mp3",
|
133 |
+
],
|
134 |
+
[
|
135 |
+
"90s rock song with electric guitar and heavy drums",
|
136 |
+
None,
|
137 |
+
],
|
138 |
+
[
|
139 |
+
"a light and cheerly EDM track, with syncopated drums, aery pads, and strong emotions bpm: 130",
|
140 |
+
"./assets/bach.mp3",
|
141 |
+
],
|
142 |
+
[
|
143 |
+
"lofi slow bpm electro chill with organic samples",
|
144 |
+
None,
|
145 |
+
],
|
146 |
+
],
|
147 |
+
inputs=[text, melody],
|
148 |
+
outputs=[output]
|
149 |
+
)
|
150 |
+
gr.Markdown("""
|
151 |
+
### More details
|
152 |
+
|
153 |
+
The model will generate 12 seconds of audio based on the description you provided.
|
154 |
+
You can optionaly provide a reference audio from which a broad melody will be extracted.
|
155 |
+
The model will then try to follow both the description and melody provided.
|
156 |
+
All samples are generated with the `melody` model.
|
157 |
+
|
158 |
+
You can also use your own GPU or a Google Colab by following the instructions on our repo.
|
159 |
+
|
160 |
+
See [github.com/facebookresearch/audiocraft](https://github.com/facebookresearch/audiocraft)
|
161 |
+
for more details.
|
162 |
+
""")
|
163 |
+
|
164 |
+
# Show the interface
|
165 |
+
launch_kwargs = {}
|
166 |
+
username = kwargs.get('username')
|
167 |
+
password = kwargs.get('password')
|
168 |
+
server_port = kwargs.get('server_port', 0)
|
169 |
+
inbrowser = kwargs.get('inbrowser', False)
|
170 |
+
share = kwargs.get('share', False)
|
171 |
+
server_name = kwargs.get('listen')
|
172 |
+
|
173 |
+
launch_kwargs['server_name'] = server_name
|
174 |
+
|
175 |
+
if username and password:
|
176 |
+
launch_kwargs['auth'] = (username, password)
|
177 |
+
if server_port > 0:
|
178 |
+
launch_kwargs['server_port'] = server_port
|
179 |
+
if inbrowser:
|
180 |
+
launch_kwargs['inbrowser'] = inbrowser
|
181 |
+
if share:
|
182 |
+
launch_kwargs['share'] = share
|
183 |
+
demo.queue(max_size=8 * 4).launch(**launch_kwargs)
|
184 |
+
|
185 |
+
|
186 |
+
if __name__ == "__main__":
|
187 |
+
parser = argparse.ArgumentParser()
|
188 |
+
parser.add_argument(
|
189 |
+
'--listen',
|
190 |
+
type=str,
|
191 |
+
default='0.0.0.0',
|
192 |
+
help='IP to listen on for connections to Gradio',
|
193 |
+
)
|
194 |
+
parser.add_argument(
|
195 |
+
'--username', type=str, default='', help='Username for authentication'
|
196 |
+
)
|
197 |
+
parser.add_argument(
|
198 |
+
'--password', type=str, default='', help='Password for authentication'
|
199 |
+
)
|
200 |
+
parser.add_argument(
|
201 |
+
'--server_port',
|
202 |
+
type=int,
|
203 |
+
default=0,
|
204 |
+
help='Port to run the server listener on',
|
205 |
+
)
|
206 |
+
parser.add_argument(
|
207 |
+
'--inbrowser', action='store_true', help='Open in browser'
|
208 |
+
)
|
209 |
+
parser.add_argument(
|
210 |
+
'--share', action='store_true', help='Share the gradio UI'
|
211 |
+
)
|
212 |
+
|
213 |
+
args = parser.parse_args()
|
214 |
+
|
215 |
+
ui(
|
216 |
+
username=args.username,
|
217 |
+
password=args.password,
|
218 |
+
inbrowser=args.inbrowser,
|
219 |
+
server_port=args.server_port,
|
220 |
+
share=args.share,
|
221 |
+
listen=args.listen
|
222 |
+
)
|
assets/a_duck_quacking_as_birds_chirp_and_a_pigeon_cooing.mp3
DELETED
Binary file (15.2 kB)
|
|
assets/sirens_and_a_humming_engine_approach_and_pass.mp3
DELETED
Binary file (15.2 kB)
|
|
audiocraft/__init__.py
CHANGED
@@ -3,24 +3,8 @@
|
|
3 |
#
|
4 |
# This source code is licensed under the license found in the
|
5 |
# LICENSE file in the root directory of this source tree.
|
6 |
-
"""
|
7 |
-
AudioCraft is a general framework for training audio generative models.
|
8 |
-
At the moment we provide the training code for:
|
9 |
-
|
10 |
-
- [MusicGen](https://arxiv.org/abs/2306.05284), a state-of-the-art
|
11 |
-
text-to-music and melody+text autoregressive generative model.
|
12 |
-
For the solver, see `audiocraft.solvers.musicgen.MusicGenSolver`, and for the model,
|
13 |
-
`audiocraft.models.musicgen.MusicGen`.
|
14 |
-
- [AudioGen](https://arxiv.org/abs/2209.15352), a state-of-the-art
|
15 |
-
text-to-general-audio generative model.
|
16 |
-
- [EnCodec](https://arxiv.org/abs/2210.13438), efficient and high fidelity
|
17 |
-
neural audio codec which provides an excellent tokenizer for autoregressive language models.
|
18 |
-
See `audiocraft.solvers.compression.CompressionSolver`, and `audiocraft.models.encodec.EncodecModel`.
|
19 |
-
- [MultiBandDiffusion](TODO), alternative diffusion-based decoder compatible with EnCodec that
|
20 |
-
improves the perceived quality and reduces the artifacts coming from adversarial decoders.
|
21 |
-
"""
|
22 |
|
23 |
# flake8: noqa
|
24 |
from . import data, modules, models
|
25 |
|
26 |
-
__version__ = '
|
|
|
3 |
#
|
4 |
# This source code is licensed under the license found in the
|
5 |
# LICENSE file in the root directory of this source tree.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
|
7 |
# flake8: noqa
|
8 |
from . import data, modules, models
|
9 |
|
10 |
+
__version__ = '0.0.2a1'
|
audiocraft/adversarial/__init__.py
DELETED
@@ -1,22 +0,0 @@
|
|
1 |
-
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
2 |
-
# All rights reserved.
|
3 |
-
#
|
4 |
-
# This source code is licensed under the license found in the
|
5 |
-
# LICENSE file in the root directory of this source tree.
|
6 |
-
"""Adversarial losses and discriminator architectures."""
|
7 |
-
|
8 |
-
# flake8: noqa
|
9 |
-
from .discriminators import (
|
10 |
-
MultiPeriodDiscriminator,
|
11 |
-
MultiScaleDiscriminator,
|
12 |
-
MultiScaleSTFTDiscriminator
|
13 |
-
)
|
14 |
-
from .losses import (
|
15 |
-
AdversarialLoss,
|
16 |
-
AdvLossType,
|
17 |
-
get_adv_criterion,
|
18 |
-
get_fake_criterion,
|
19 |
-
get_real_criterion,
|
20 |
-
FeatLossType,
|
21 |
-
FeatureMatchingLoss
|
22 |
-
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|