Sarvam TTS plugin guide | LiveKit Documentation

Available in

Python

Node.js

Overview

This plugin allows you to use Sarvam as a TTS provider for your voice agents.

Installation

Install the plugin:

uv add "livekit-agents[sarvam]~=1.4"

pnpm add @livekit/agents-plugin-sarvam@1.x

Authentication

The Sarvam plugin requires a Sarvam API key.

Set SARVAM_API_KEY in your .env file.

Usage

Use Sarvam TTS within an AgentSession or as a standalone speech generator. For example, you can use this TTS in the Voice AI quickstart.

from livekit.plugins import sarvam

session = AgentSession(
   tts=sarvam.TTS(
      target_language_code="hi-IN",
      model="bulbul:v3",
      speaker="shubh",
      pace=1.0,
      temperature=0.6,
      output_audio_bitrate="128k",
      min_buffer_size=50,
      max_chunk_length=150,
   )
   # ... llm, stt, etc.
)

import * as sarvam from '@livekit/agents-plugin-sarvam';

const session = new voice.AgentSession({
    tts: new sarvam.TTS({
        targetLanguageCode: "hi-IN",
        model: "bulbul:v3",
        speaker: "shubh",
        pace: 1.0,
        temperature: 0.6,
    }),
    // ... llm, stt, etc.
});

Parameters

This section describes some of the available parameters. See the plugin reference links in the Additional resources section for a complete list of all available parameters.

target_language_codestringRequired

BCP-47 language code for supported Indian languages. For example: hi-IN for Hindi, en-IN for Indian English. See documentation for a complete list of supported languages.

In Node.js this parameter is called targetLanguageCode.

modelstringOptionalDefault: bulbul:v2

The Sarvam TTS model to use. Valid values are:

bulbul:v2
bulbul:v3-beta
bulbul:v3

speakerstringOptionalDefault: varies by model

Voice to use for synthesis. Default depends on the selected model:

anushka for bulbul:v2
shubh for bulbul:v3-beta and bulbul:v3

Speakers are validated for model compatibility.

pitchfloatOptionalDefault: 0.0

Voice pitch adjustment. Valid range: -20.0 to 20.0. Included in synthesis payload for bulbul:v2.

pacefloatOptionalDefault: 1.0

Speech rate multiplier. Valid range: 0.5 to 2.0.

temperaturefloatOptionalDefault: 0.6

Controls output randomness. Valid range: 0.01 to 1.0. Only valid if model is bulbul:v3 or bulbul:v3-beta. This value is ignored for bulbul:v2.

loudnessfloatOptionalDefault: 1.0

Volume multiplier. Valid range: 0.5 to 2.0. Included in synthesis payload for bulbul:v2.

enable_preprocessingbooleanOptionalDefault: false

Controls whether normalization of English words and numeric entities (for example, numbers and dates) is performed. Set to true for better handling of mixed-language text.

Only valid if model is bulbul:v2. This value is ignored for other models.

In Node.js this parameter is called enablePreprocessing.

output_audio_bitratestringOptionalDefault: 128k

Output audio bitrate. Allowed values: 32k, 64k, 96k, 128k, 192k.

Only available in the Python plugin.

min_buffer_sizeintegerOptionalDefault: 50

Minimum character length that triggers buffer flushing for TTS model processing. Valid range: 30 to 200.

Only available in the Python plugin.

max_chunk_lengthintegerOptionalDefault: 150

Maximum length for sentence splitting (adjust based on content length). Valid range: 50 to 500.

Only available in the Python plugin.

speech_sample_rateintOptionalDefault: 22050

Output sample rate in Hz. Supported values: 8000, 16000, 22050, 24000.

In Node.js this parameter is called sampleRate.

Additional resources

The following resources provide more information about using Sarvam with LiveKit Agents.

Python

PyPI Reference GitHub

Node.js

NPM Reference GitHub

Sarvam docs

Sarvam's full docs site.

Voice AI quickstart

Get started with LiveKit Agents and Sarvam.

Sarvam STT

Guide to the Sarvam STT plugin with LiveKit Agents.