speech-to-speech

conda create --name <env_name> python=3.7
conda activate <env_name>

Install speech-to-speech dependencies

git clone https://github.com/Open-Speech-EkStep/speech-to-speech.git
cd speech-to-speech
bash s2s_installs.sh
pip install -r requirements.txt

ASR models

Update ASR model paths in: infer.sh

TTS models

Update vakyansh-TTS model paths in: tts_inference.py

Run

To run a Gradio interface for Speech to Speech conversion from English to Hindi, simply run:

bash infer.sh