Speech-to-Speech POC JioStar

<aside> 💡

POC for JioStar
https://github.com/hanspaa2017108/speech-to-speech </aside>
Led the research and development of speech-to-speech for JioStar
The core idea is to convert English audio to Indian regional languages
Researched multiple ways to built the speech-to-speech pipeline, prepared technical report explaining which is better focusing on three parameters — Latency, Speed, Output Accuracy
This POC is built on top of OpenAI’s Realtime API

English Input-

Hindi Output-

Marathi Output-

Punjabi Output-

Gujarati Output-