livekit-plugins-gnani 0.4.0

LiveKit Agents plugin for Gnani Vachana speech AI — STT & TTS for Indian languages

livekit-plugins-gnani

LiveKit Agents plugin for Gnani Vachana — high-accuracy Speech-to-Text and low-latency Text-to-Speech for Indian languages.

Vachana is a production-ready speech AI platform by Gnani.ai supporting 10+ Indian languages with 6 voices, real-time streaming, multilingual transcription, and code-switching capabilities.

Installation

pip install livekit-plugins-gnani

This will also install the gnani-vachana core SDK as a dependency.

Prerequisites

You need a Gnani API key. Email [e-mail] to get started — all new accounts receive free credits, no credit card required.

Set your credentials as environment variables:

export GNANI_API_KEY="your-api-key"

# For REST STT only (optional):
export GNANI_ORGANIZATION_ID="your-org-id"
export GNANI_USER_ID="your-user-id"

Quick Start

Speech-to-Text

from livekit.plugins.gnani import STT

stt = STT(language="hi-IN")

# Use with a LiveKit voice agent pipeline

Text-to-Speech

from livekit.plugins.gnani import TTS

# REST (default) — single-request batch synthesis
tts = TTS(voice="Karan")

# SSE — streaming via Server-Sent Events (lower latency)
tts = TTS(voice="Karan", synthesize_method="sse")

# WebSocket — real-time streaming via stream() (lowest latency)
tts = TTS(voice="Karan", synthesize_method="websocket")

All three modes work with the standard LiveKit voice agent pipeline. The synthesize_method controls which endpoint synthesize() uses. The stream() method always uses WebSocket regardless of this setting.

Features

STT

• Batch recognition — REST API (POST /stt/v3) for file-based transcription
• Real-time streaming — WebSocket API for live audio transcription with VAD
• 23 Indian languages — Assamese, Bengali, Bodo, Dogri, English (India), Gujarati, Hindi, Kannada, Kashmiri, Konkani, Maithili, Malayalam, Manipuri, Marathi, Nepali, Odia, Punjabi, Sanskrit, Santhali, Sindhi, Tamil, Telugu, Urdu
• Code-switching — Hinglish (en-hi-IN-latn) and Hindi-English mixed (en-hi-in-cm) for streaming
• Sample rates — 8 kHz and 16 kHz

TTS

• REST synthesis — single-request batch audio generation (synthesize_method="rest")
• SSE streaming — lower-latency chunked synthesis via Server-Sent Events (synthesize_method="sse")
• WebSocket streaming — lowest-latency real-time synthesis via stream() or synthesize_method="websocket"
• 6 voices — Karan, Simran, Nara, Riya, Viraj, Raju
• Model — vachana-voice-v3 with voice cloning support
• Configurable output — sample rate (8000–44100), encoding (linear_pcm, oggopus), container (raw, mp3, wav, mulaw, ogg)

Supported Languages

STT Languages (Speech-to-Text) — 10 languages

STT uses BCP-47 locale codes (e.g. hi-IN):

Language	Code
English (India)	en-IN
Hindi	hi-IN
Gujarati	gu-IN
Tamil	ta-IN
Kannada	kn-IN
Telugu	te-IN
Marathi	mr-IN
Bengali	bn-IN
Malayalam	ml-IN
Punjabi	pa-IN

Plus streaming-only experimental: en-hi-IN-latn (Hinglish), en-hi-in-cm (code-mixed).

---

TTS Languages (Text-to-Speech) — 10 languages

TTS uses ISO 639 language codes (e.g. hi, bn). Pass these via the language parameter.

Language	Code
Assamese	as
Bengali	bn
English	en
Hindi	hi
Kannada	kn
Malayalam	ml
Marathi	mr
Odia	or
Tamil	ta
Telugu	te

Available Voices

Voice	ID	Gender	Description
Karan	Karan	Male	Bold, Trustworthy
Simran	Simran	Female	Confident, Bright
Nara	Nara	Female	Gentle, Expressive
Riya	Riya	Female	Cheerful, Energetic
Viraj	Viraj	Male	Commanding, Dynamic
Raju	Raju	Male	Grounded, Conversational

Architecture

gnani-vachana (>=0.4.0)   <- Core SDK (REST, WebSocket, SSE clients, voice/language constants)
    |
livekit-plugins-gnani     <- This package (LiveKit Agents adapter)

This plugin is a thin adapter that wraps the gnani-vachana SDK into LiveKit's stt.STT and tts.TTS base classes. Voice lists, language constants, and model definitions are shared with the core SDK.

Documentation

• Vachana API Docs
• LiveKit Agents Docs
• gnani-vachana SDK

License

Apache 2.0 — see LICENSE.