Freedom251 › Transcription Software
Sarvam AI Speech to Text
What Sarvam AI Speech to Text is
Sarvam AI Speech to Text converts audio into text through a browser playground and an API. Saaras v3 supports 23 languages—all 22 scheduled Indian languages plus English—and can detect language automatically. Output modes include transcription, translation, transliteration, verbatim, and codemix, with word-level timestamps by default. The Batch API identifies speakers and assigns stable IDs to words, including in overlapping speech. Streaming runs over WebSocket, with the product page stating sub-150ms time to first token in Fast mode. The page lists use cases including voice agents, post-call analytics, captions, meeting transcription, and dictation. REST handles files under 30 seconds; Batch handles files up to one hour, while pricing details also list batch plans for files up to two hours. The free browser playground and signup credits offer ways to begin; API rates listed are ₹30.00 per hour for Real-time, Streaming, and Batch, and ₹45.00 per hour for Batch with diarization. Sarvam says API inputs are not used for training and are deleted after processing by default. The service is headquartered in Bengaluru, India.
Who it suits
It suits developers and teams building speech features or processing audio for analytics, captions, meetings, or dictation. The API, streaming, batch, and self-hosted options may fit different deployment needs.
What it does well
- Supports 23 languages with automatic language detection.
- Offers transcription, translation, transliteration, verbatim, and codemix modes.
- Batch diarization identifies speakers, including overlapping speech.
- Lists official Python and JavaScript SDKs.
- Signup includes credits that do not expire.
What to know first
- REST API handles files under 30 seconds.
- API processing rates are billed by the hour.
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Real-time | Speech-to-text API | ₹30 a month |
| Streaming | Streaming speech-to-text | ₹30 a month |
| Batch | Up to 2 hours per file | ₹30 a month |
| Batch with diarization | Up to 2 hours per file | ₹45 a month |
| Enterprise | Custom pricing for high-volume or production-critical workloads. · Custom limits, throughput, pricing, and support · Higher rate limits · Custom concurrency · Volume discounts · Dedicated support and SLAs | On request |
| Starter (Pay as you go) | No monthly commitment; credits can be topped up as needed and never expire. · Speech to text: Real-time ₹30.00/hour · Streaming ₹30.00/hour · Batch ₹30.00/hour · Batch with diarization ₹45.00/hour | On request |
As sarvam.ai lists them, read 20 Sept 2026. The maker's own page has the final word.
Verdict
Sarvam offers multiple output modes and processing routes, with speaker identification available in Batch and streaming over WebSocket. Before choosing a batch workflow, check the differing file limits stated for the API and pricing plans.
Facts from Sarvam AI Speech to Text's own pages
- API access
- Yes · sarvam.ai, 20 Sept 2026
- API limits
- The REST API handles files under 30 seconds, while Batch handles files up to one hour; Streaming transcribes in real time over WebSocket. · sarvam.ai, 8 Oct 2026
- Code-mixing and noise
- Sarvam says Saaras v3 handles mixed-language speech, noisy telephony, and diverse accents, and was trained on over one million hours of real-world audio. · sarvam.ai, 8 Oct 2026
- Compliance
- The product page lists SOC 2 Type II, ISO 27001, and DPDP compliance, along with role-based access and audit trails. · sarvam.ai, 8 Oct 2026
- Credits
- The pricing page says signup includes free credits, credits are universal across supported APIs, and unused credits do not expire. · sarvam.ai, 8 Oct 2026
- Deployment
- Sarvam says regulated workloads can use VPC or on-premise deployment; its self-hosted guide describes deploying a Saaras model on Amazon SageMaker in the customer’s AWS account and VPC. · docs.sarvam.ai, 8 Oct 2026
- Free plan
- No · sarvam.ai, 20 Sept 2026
- Headquarters
- Bengaluru, Karnataka, India · sarvam.ai, 28 Sept 2026
- Integrations
- Sarvam lists LiveKit, Pipecat, n8n, Vercel AI SDK, and LangChain integrations, plus official Python and JavaScript SDKs. · docs.sarvam.ai, 8 Oct 2026
- Languages supported
- 23 · sarvam.ai, 20 Sept 2026
- Languages
- Saaras v3 supports all 22 scheduled Indian languages plus English, with automatic language detection. · sarvam.ai, 8 Oct 2026
- Output modes
- The page describes transcribe, translate, transliteration, verbatim, and codemix output modes, with word-level timestamps by default. · sarvam.ai, 8 Oct 2026
- Product
- Sarvam Speech to Text converts audio into text and offers a free browser playground and an API. · sarvam.ai, 8 Oct 2026
- Security
- Sarvam states API inputs are not used for model training, data is deleted after processing by default, and processing occurs within India without cross-border transfers. · sarvam.ai, 8 Oct 2026
- Speaker diarization
- The Batch API identifies speakers and tags each word with a stable speaker ID, including for overlapping speech. · sarvam.ai, 8 Oct 2026
- Speaker identification
- Yes · sarvam.ai, 20 Sept 2026
- Streaming
- The product page states sub-150ms time to first token in Fast mode and four streaming modes via WebSocket. · sarvam.ai, 8 Oct 2026
- Support
- Sarvam’s API documentation lists support by email at [email protected] and a Discord community. · docs.sarvam.ai, 8 Oct 2026
- Timestamp support
- Yes · sarvam.ai, 20 Sept 2026
- Use cases
- The product page names voice agents, post-call analytics, subtitling and captioning, meeting transcription, and voice dictation as use cases. · sarvam.ai, 8 Oct 2026
Where Sarvam AI Speech to Text stands on Freedom251
- Transcription Software#22 of 45