Recommendations for Indian language TTS, STT, and LLM pipeline

Getting Started
1

This question originally came up in our Slack community and the thread has been consolidated here for long-term reference.

What’s the best TTS, STT, and LLM pipeline for Indian languages?

Currently I’m using Sarvam for STT and TTS, and GPT-4o for LLM, but there’s latency on the LLM side that sometimes reaches 2-3 seconds.

2

For TTS and STT in Indian languages, you can try:

  • smallest.ai - Supports Indian languages
  • AI4Bharat - Available on HuggingFace, good for regional languages
  • Sarvam AI - Also recommended for Indian language support