change readme
Browse files
README.md
CHANGED
@@ -16,9 +16,11 @@ task_categories:
|
|
16 |
extra_gated_prompt: "When using this dataset to download LibriTTS, you agree to the terms on https://www.openslr.org"
|
17 |
---
|
18 |
|
19 |
-
|
20 |
|
21 |
-
|
|
|
|
|
22 |
You need to run ``pip install alignments phones`` before using this dataset.
|
23 |
When running this the first time, it can take an hour or two, but subsequent runs will be lightning fast.
|
24 |
|
@@ -26,6 +28,8 @@ When running this the first time, it can take an hour or two, but subsequent run
|
|
26 |
- ``pip install alignments phones`` **(required)**
|
27 |
- ``pip install speech-collator`` (optional)
|
28 |
|
|
|
|
|
29 |
## Example Item
|
30 |
|
31 |
```json
|
|
|
16 |
extra_gated_prompt: "When using this dataset to download LibriTTS, you agree to the terms on https://www.openslr.org"
|
17 |
---
|
18 |
|
19 |
+
> This dataset is identical to **[cdminix/libritts-aligned](https://huggingface.co/datasets/cdminix/libritts-aligned)** except it uses the newly released LibriTTS-R corpus. Please cite **[Y. Koizumi, et al., "LibriTTS-R: Restoration of a Large-Scale Multi-Speaker TTS Corpus", Interspeech 2023](https://google.github.io/df-conformer/librittsr/)**
|
20 |
|
21 |
+
# Dataset Card for LibriTTS-R with Forced Alignments (and Measures)
|
22 |
+
|
23 |
+
This dataset downloads LibriTTS-R and preprocesses it on your machine to create alignments using [montreal forced aligner](https://montreal-forced-aligner.readthedocs.io/en/latest/).
|
24 |
You need to run ``pip install alignments phones`` before using this dataset.
|
25 |
When running this the first time, it can take an hour or two, but subsequent runs will be lightning fast.
|
26 |
|
|
|
28 |
- ``pip install alignments phones`` **(required)**
|
29 |
- ``pip install speech-collator`` (optional)
|
30 |
|
31 |
+
*Note: version >=0.0.14 of alignments is required for this corpus*
|
32 |
+
|
33 |
## Example Item
|
34 |
|
35 |
```json
|