Home
Overview
Speech to Text
Getting StartedPre-recorded AudioStreaming AudioModels & Languages
Text to Speech
Getting StartedBatch TTSStreaming TTSVoices & Languages
Resources
API ReferencePython SDK

v1.1 · Updated Apr 2025

Home
Overview
Speech to Text
Getting StartedPre-recorded AudioStreaming AudioModels & Languages
Text to Speech
Getting StartedBatch TTSStreaming TTSVoices & Languages
Resources
API ReferencePython SDK
HomeSpeech to TextText to SpeechAPI Reference
PlaygroundConsole
Home

Developer Documentation

Build with ShunyaLabs

Speech AI APIs built for Indian languages. High-accuracy speech-to-text, natural text-to-speech, and a full suite of NLP features — all in one platform.


Products

Speech to Text

High-accuracy, low-latency speech recognition for 55+ Indian languages. Batch transcription and real-time streaming with built-in NLP.

55 languagesReal-time streamingNLP enrichment
Get started

Text to Speech

Natural speech synthesis with 46 voices across 23 Indic languages. Batch and streaming modes with expression styles and voice cloning.

46 voices11 expression stylesVoice cloning
Get started

API Reference

Complete endpoint reference for both STT and TTS APIs. REST endpoints, WebSocket streaming, and speaker management.

REST APIWebSocketOpenAI-compatible
Get started

Quick links

STT Quickstart

Transcribe your first audio file in 5 minutes.

TTS Quickstart

Generate your first audio from text.

Python SDK

Install and configure the ShunyaLabs SDK.

Models Overview

Compare zero-indic, zero-med, zero-codeswitch, and zero-universal.