Update README.md
Browse filesAdded the speaker embeddings model reference.
README.md
CHANGED
@@ -4,14 +4,15 @@ base_model: microsoft/speecht5_tts
|
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
-
- voxpopuli
|
8 |
model-index:
|
9 |
- name: speecht5_tts-ft-voxpopuli-it
|
10 |
results: []
|
|
|
|
|
11 |
---
|
12 |
|
13 |
-
|
14 |
-
should probably proofread and complete it, then remove this comment. -->
|
15 |
|
16 |
# speecht5_tts-ft-voxpopuli-it
|
17 |
|
@@ -21,7 +22,7 @@ It achieves the following results on the evaluation set:
|
|
21 |
|
22 |
## Model description
|
23 |
|
24 |
-
|
25 |
|
26 |
## Intended uses & limitations
|
27 |
|
@@ -29,7 +30,7 @@ More information needed
|
|
29 |
|
30 |
## Training and evaluation data
|
31 |
|
32 |
-
|
33 |
|
34 |
## Training procedure
|
35 |
|
@@ -61,4 +62,4 @@ The following hyperparameters were used during training:
|
|
61 |
- Transformers 4.33.0
|
62 |
- Pytorch 1.12.1+cu116
|
63 |
- Datasets 2.14.4
|
64 |
-
- Tokenizers 0.12.1
|
|
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
+
- facebook/voxpopuli
|
8 |
model-index:
|
9 |
- name: speecht5_tts-ft-voxpopuli-it
|
10 |
results: []
|
11 |
+
language:
|
12 |
+
- it
|
13 |
---
|
14 |
|
15 |
+
|
|
|
16 |
|
17 |
# speecht5_tts-ft-voxpopuli-it
|
18 |
|
|
|
22 |
|
23 |
## Model description
|
24 |
|
25 |
+
It uses the speaker embedding model speechbrain/spkrec-xvect-voxceleb
|
26 |
|
27 |
## Intended uses & limitations
|
28 |
|
|
|
30 |
|
31 |
## Training and evaluation data
|
32 |
|
33 |
+
test_size=0.15
|
34 |
|
35 |
## Training procedure
|
36 |
|
|
|
62 |
- Transformers 4.33.0
|
63 |
- Pytorch 1.12.1+cu116
|
64 |
- Datasets 2.14.4
|
65 |
+
- Tokenizers 0.12.1
|