iamilyasedunov / asr_mvp

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Prepare

python3.8 -m venv venv_exp
source venv_exp/bin/activate
python3.8 -m pip install --upgrade git+https://github.com/huggingface/transformers.git accelerate torchaudio

Modify manual

usage: main.py modify [-h] [--speed SPEED] [--volume VOLUME] input_file output_file

positional arguments:
  input_file       Path to input WAV file
  output_file      Path to output modified WAV file

optional arguments:
  -h, --help       show this help message and exit
  --speed SPEED    Speed factor (default: 1.0)
  --volume VOLUME  Volume change in dB (default: 0.0)

Ex:

python3.8 main.py modify sample.wav sample_edited.wav --volume 10.0 --speed 0.5

Transcribe manual

usage: main.py transcribe [-h] [--lang {russian,english}] input_file

positional arguments:
  input_file            Path to input WAV file

optional arguments:
  -h, --help            show this help message and exit
  --lang {russian,english}
                        Language of the model (default: english)

Ex:

python3.8 main.py transcribe sample.wav --lang russian

About


Languages

Language:Python 100.0%