https://github.com/mlcommons/peoples-speech
The People’s Speech Dataset
https://github.com/mlcommons/peoples-speech
Last synced: 11 months ago
JSON representation
The People’s Speech Dataset
- Host: GitHub
- URL: https://github.com/mlcommons/peoples-speech
- Owner: mlcommons
- License: apache-2.0
- Created: 2021-02-11T23:16:38.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2024-01-11T19:14:32.000Z (over 2 years ago)
- Last Synced: 2025-06-09T14:12:56.820Z (about 1 year ago)
- Language: Jupyter Notebook
- Homepage: https://mlcommons.org/en/peoples-speech/
- Size: 141 MB
- Stars: 104
- Watchers: 15
- Forks: 12
- Open Issues: 37
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE.md
- Codeowners: .github/CODEOWNERS
Awesome Lists containing this project
- Awesome-SpeechLM-Survey - People dataset - Training | 30k | 2021 | (Popular Training Datasets / Mixed Tokenizers)
README
# People's Speech Data Pipelines
Installation
```
# libprotobuf-dev is an onnx dependency, transitively brought in by nemo.
sudo apt-get install git-lfs sox ffmpeg
# Set up a virtual environment of some sort
pip install numpy Cython
python setup.py develop
cp galvasr2/*.jar $(python -c "import pyspark; print(pyspark.__path__[0])")/jars
```
Run forced alignment pipeline.
```
python galvasr2/align/spark/align_cuda_decoder.py --stage=0
```