Sarvam AI Saaras Realtime STT
Indian-language focused STT with a 2026 Realtime WebSocket API (saaras:v3-realtime default, saaras:v4) priced in rupees per hour.
Overview
Best for: Voice agents and transcription for Indian languages and Hinglish.
At a glance
Priced in INR: INR 30/hr (INR 45/hr with diarization), converted at about INR 85 per USD. Free credit INR 100. Indian languages plus English, exact list not captured. Pricing does not separate streaming from batch.
encoding and sample_rate parameters on the realtime channel (values not captured).
n/a
Indian languages plus English (exact list not captured).
Not captured.
India (not re-verified).
Not re-verified.
Features
- realtime transcription
- speech-to-text with translation
- diarization (priced variant)
Pricing
| What | Price | Unit |
|---|---|---|
| Speech to Text | INR 30 | per hour, billed per second |
| STT with diarization | INR 45 | per hour |
| STT and translate | INR 30 | per hour |
INR 30-45/hr converted at about INR 85 per USD (third-party FX note, July 2026); currency risk applies.
Free tier: INR 100 credits for new users.
Source: docs.sarvam.ai
Setup
- Get an API subscription key from the Sarvam dashboard.
- Connect to wss://api.sarvam.ai/speech-to-text-realtime/ws with model, language_code, encoding and sample_rate parameters.
- Server: send header Api-Subscription-Key; browser: use WebSocket subprotocol 'api-subscription-key.<key>' (exposes the key, so prefer a server relay).
Endpoint
wss://api.sarvam.ai/speech-to-text-realtime/wsAuthentication
Api-Subscription-Key header, or subprotocol api-subscription-key.<key> in browsers.
Warnings
Browser auth exposes the key
The documented browser method puts the subscription key in the WebSocket subprotocol; no short-lived token mechanism was found. Relay through your server.
Two streaming APIs
The newer Realtime API (saaras:v3-realtime) supersedes the legacy streaming WebSocket for new work; message schemas differ, so do not mix examples.
Rupee pricing
Prices are published only in INR; USD cost moves with the exchange rate.
Diarization raises price 50%
STT with diarization is INR 45/hr vs INR 30/hr.
Plus 12 warnings that apply to all speech-to-text, live APIs. See category warnings.
Limits
- See Credits and Rate Limits page (not captured).
Models and products
| Name | Status |
|---|---|
| saaras:v3-realtime | GA (default on Realtime API) |
| saaras:v4 | GA |
| saaras:v3 legacy streaming WebSocket | Legacy |
Docs and sources
Docs
Sources used
- docs.sarvam.ai/api/pricing.md
- docs.sarvam.ai/api-reference/speech-to-text/transcribe/realtime/ws.md
- docs.sarvam.ai/changelog/2026/8/1
- usagepricing.com/blueprint/sarvam-ai
Message schema (no snippet given), language list, rate limits, latency, and whether the realtime API has a distinct price.