Transcriber
Sarvam AI (STT) logo

Sarvam AI (STT)

Add Sarvam Saaras speech-to-text to Bolna voice agents for accurate Indian language transcription. Handles regional accents and Hinglish code-switching on live phone calls.

Book a DemoBolna Docs
stttranscriberindiasarvamhindiregional-languagestamiltelugukannada
Official Documentation
At a glance

How Sarvam AI (STT) fits in the stack

Best for

Turning noisy phone audio into text the agent can reason over.

Use this layer when

Recognition accuracy, language coverage, and latency matter most.

Connects to

Telephony audio upstream and your LLM downstream.

Voice Stack

Speech-to-text is the listening layer

Speech-to-text is the listening layer

This provider converts raw audio into text in real time. It shapes how accurately the agent hears the caller and how natural the back-and-forth feels.

TelephonyPhone Network
→
STTListener
→
LLMReasoning
→
TTSVoice
→
ToolsActions

This page focuses on where Sarvam AI (STT) fits in a production voice stack. For full setup steps, credentials, and API details, use the documentation link above.

Overview

Sarvam AI builds speech models trained on Indian languages from the ground up, rather than adapting an English-first model after the fact. Its Saaras speech-to-text family understands heavy regional accents, code-mixed speech, and the mid-sentence language switching that is normal on Indian phone calls.

With Bolna's Sarvam integration, your voice agents transcribe caller audio in real time in Indian languages and Indian English, keeping conversations responsive instead of laggy.

Models

Bolna supports these Saaras models:

  • saaras:v4 - Sarvam's latest. Transcribes in the original spoken language with automatic language detection, and adds global English accents alongside Indian English. Recommended for new agents.
  • saaras:v3 - Sarvam's current default model, transcribing in the original spoken language.

Supported Languages

Saaras covers 22 Indian languages plus English:

Assamese, Bengali, Bodo, Dogri, English (India), Gujarati, Hindi, Kannada, Kashmiri, Konkani, Maithili, Malayalam, Manipuri, Marathi, Nepali, Odia, Punjabi, Sanskrit, Santali, Sindhi, Tamil, Telugu, and Urdu.

See the Bolna docs for the language codes you can set on an agent today.

Features & Use Cases

Built for Indian Speech
Trained directly on Indian language data, so regional accents and dialects transcribe accurately rather than approximately.

Code-Switching Ready
Handles callers who mix English with a regional language mid-sentence, like Hinglish or Tanglish, without needing to reconfigure the agent.

Real-Time Streaming
Streaming transcription over WebSocket keeps latency low enough for natural back-and-forth conversation.

Automatic Language Detection
Saaras models can identify the spoken language rather than requiring it to be declared up front.

Use Case: All-India Customer Support
Run support lines that recognise the caller's language and respond in it, without routing them through a menu first.

Use Case: Reaching Rural Markets
Connect with customers outside metros by transcribing local dialects accurately enough to act on.

Use Case: Regional Banking Services
Meet requirements for financial services in local languages while keeping the experience conversational.

Related

Other providers in this layer

Transcriber

AssemblyAI

Add AssemblyAI Universal-3.6 Pro speech-to-text to Bolna voice agents. Built for real calls, it covers 32 languages with code-switching and gets short answers like "yes" right.

Transcriber

Deepgram

Plug Deepgram speech-to-text into Bolna for fast, accurate phone call transcription. Supports low latency streaming, high accuracy, and custom vocabulary.

Transcriber

Smallest AI (Pulse)

Add Smallest AI Pulse speech-to-text to Bolna for ultra-fast, low-latency phone call transcription. Built for real-time voice agent conversations.

Browse this layer

Keep exploring the voice stack

Browse

Speech-to-Text

Speech-to-text converts what callers say into text that your LLM can process. Transcription accuracy and latency directly affect how natural a conversation feels. Bolna supports streaming STT providers optimized for telephony audio, including specialized models for Indian languages.

Browse

Telephony

Telephony providers connect your voice agents to the phone network so they can make and receive real calls. Bolna supports managed integrations with major carriers as well as bring-your-own-carrier via SIP trunking, giving you full control over call routing, number provisioning, and cost.

Browse

Large Language Models

The LLM is the brain of your voice agent. It understands what callers say and decides how to respond. Bolna lets you swap between models like GPT-4o, Claude, and DeepSeek without changing your agent configuration, so you can optimize for speed, cost, or reasoning depth.

Browse

Text-to-Speech

Text-to-speech turns your agent responses into spoken audio. Voice quality shapes how callers perceive your brand. Flat, robotic speech kills trust while natural, expressive voices build it. Bolna integrates with the fastest TTS providers so responses sound human and arrive without awkward pauses.

Browse

Tools & Workflows

Tools let your voice agents take action during a call, not just talk. Book a calendar slot, look up an order in Shopify, push a lead into your CRM, or trigger a multi-step automation in Zapier. These integrations turn voice agents from answering machines into workflow engines.

See where Sarvam AI (STT) fits in your production workflow

Use the demo to walk through provider selection, stack tradeoffs, and the exact workflow you want Bolna to automate.

Book a DemoRead Bolna Docs