Speech to Text for Indian Languages

Saaras v3 automatically detects the spoken language without requiring you to specify it upfront. When a speaker switches between Hindi and English mid-sentence, or drops Tamil words into an otherwise English conversation, the model follows along without losing accuracy at language boundaries. This is not a pipeline that detects language segments and routes them to separate engines. It is a single unified model that understands multilingual Indian speech as a continuous stream.

An estimated 350 million Indians speak English as a second language, and the vast majority mix it freely with their primary language. A customer might say "Mera order ka status kya hai, I placed it last Tuesday." A doctor might dictate "Patient ko high fever hai, CBC and LFT karwao." This is normal Indian communication, and Saaras v3 transcribes it accurately because it was trained on exactly this kind of speech.

Most global ASR models are trained on clean, studio-quality audio. Real Indian audio is not clean. Call center recordings come in at 8kHz with compression artifacts. Field interviews have traffic noise. Group discussions have overlapping speakers. Mobile calls have network dropouts. Saaras v3 is trained on exactly this kind of audio, so it holds up where other models degrade.

No pre-processing or noise cancellation needed on your end. The robustness comes from a 4-stage training pipeline: pre-training on diverse acoustic conditions, supervised fine-tuning, reinforcement learning, and a post-training stage that specifically targets long-tail errors and output stability. For voice agent deployments where callers are speaking from busy streets, factory floors, or crowded homes, this is the difference between a system that works in production and one that only works in demos.

Speaker diarization answers "who said what?" in a multi-speaker recording. The Batch API identifies speakers and tags each word with a stable speaker ID. You need this for meeting transcriptions (attributing action items), call center analytics (separating agent and customer utterances), and interview recordings (distinguishing interviewer from interviewee).

Diarization works across all 22 supported languages and handles overlapping speech, speaker turns of varying length, and conversations where speakers have similar voice characteristics. The speaker labels are consistent across the entire recording, so Speaker 1 at minute 2 is the same person as Speaker 1 at minute 45.

Voice dictation (speaking instead of typing) is how a lot of people prefer to interact with devices, especially on mobile. Sarvam's speech recognition powers voice typing in Indian languages: dictate messages, emails, notes, and documents in Hindi, Tamil, Telugu, or any of the 22 supported languages. Unlike generic voice dictation that forces you into English, Sarvam outputs correct Devanagari (or whatever script you are dictating in), adds punctuation automatically, and keeps up with natural speaking speed. If you type faster by talking, or find keyboards difficult, voice dictation closes that gap.

You can convert speech to text online using the playground above, or integrate via API for production use. Sign up at sarvam.ai/try/stt-api to get your API key. Install the Python SDK with pip install sarvamai or the Node.js SDK with npm install sarvamai. Your first audio to text conversion takes under 5 minutes from signup to working output.