https://github.com/egorsmkv/w2v2-bert-aligner
Aligner for wav2vec2-bert models
https://github.com/egorsmkv/w2v2-bert-aligner
asr automatic-speech-recognition ctc ukrainian wav2vec2
Last synced: about 1 year ago
JSON representation
Aligner for wav2vec2-bert models
- Host: GitHub
- URL: https://github.com/egorsmkv/w2v2-bert-aligner
- Owner: egorsmkv
- Created: 2024-12-19T22:01:44.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2024-12-19T22:02:48.000Z (over 1 year ago)
- Last Synced: 2024-12-19T23:17:59.944Z (over 1 year ago)
- Topics: asr, automatic-speech-recognition, ctc, ukrainian, wav2vec2
- Language: Python
- Homepage:
- Size: 1.95 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Align audio and text using wav2vec2-bert models
## Community
- **Discord**: https://bit.ly/discord-uds
- Speech Recognition: https://t.me/speech_recognition_uk
- Speech Synthesis: https://t.me/speech_synthesis_uk
## Install
```bash
uv venv --python 3.12
source .venv/bin/activate
uv pip install -r requirements.txt
# in development mode
uv pip install -r requirements-dev.txt
# check/format the code
ruff check --select I --fix
ruff format
```
## Test
### Download & convert the data
```bash
wget "https://github.com/egorsmkv/cv10-uk-testset-clean/releases/download/v1.0/filtered-cv10-test.zip"
unzip filtered-cv10-test.zip
wget https://raw.githubusercontent.com/egorsmkv/cv10-uk-testset-clean/refs/heads/main/rows.csv
```
### Run aligment
```bash
python alignment.py
```