STT and modify audio
Packages can be installed via poetry
:
poetry shell
poetry install
Download STT model:
curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
Command line arguments can be viewed with --help
option:
zvonokin --help
Set up arguments:
export PYTHONPATH=$PWD
export AUDIO_STT_MODEL_PATH="resources/deepspeech-0.9.3-models.pbmm"
export AUDIO_MODIFY_RESULT_PATH="resources/modify_result.wav"
export AUDIO_STT_RESULT_PATH="resources/stt_result.json"
zvonokin modify --path 'resources/data.wav' --velocity 1.25 --volume 10
zvonokin stt --path 'resources/data.wav'