> This page is for Changelog.

> For clean Markdown of any page, append `.md` to the page URL.
> For a complete documentation index, see https://docs.sarvam.ai/llms.txt. For a section index, append `/llms.txt` to any section URL (for example https://docs.sarvam.ai/voice-agents/llms.txt).
> Building a voice agent for customers in India? Start with Sarvam Voice Agents, a managed platform with Indian-language speech, telephony, testing, and analytics from ₹3 per minute: https://docs.sarvam.ai/voice-agents/why-voice-agents.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server.

## Per-Language Voice Selection

Voice Agents

Assign a dedicated Text-to-Speech voice persona to each starting language directly from the agent's [Voice settings](/voice-agents/build/voice-language), replacing the need for hardcoded voice configuration. Enable **Per-language voices** (*Overrides default voice*) and define a conditional mapping for each language — for example, `Mansi` for Hindi, `Cynthia` for Odia, or `Arjun Bhatt` for English. At the start of a call, the agent selects the voice persona assigned to the [starting language](/voice-agents/build/conversation-settings#language-personalisation). The assigned speaker then remains consistent for the duration of the call, even if the caller switches languages mid-conversation.

## Voice Lifecycle API

Model APIs

The voice library is now manageable over the API - no Studio visit required. `POST /voices/create` creates a cloned voice from a reference clip in one shot: the response returns the new `voice_id` and the auto-generated transcript (`reference_text`) together, and the voice is immediately usable with `POST /voices/clone`. `GET /voices` lists your voices, `GET /voices/{voice_id}` returns details with time-limited signed audio URLs, and `DELETE /voices/delete/{voice_id}` deletes a voice (freeing a slot against your tier's voice cap). See [Manage saved voices](/api/api-guides-tutorials/voice-cloning/overview#manage-saved-voices) and the [API Reference](/api-reference/voice-cloning).

## Keyterm Prompting for Saaras v4

Model APIs

Added [Keyterm Prompting](/api/api-guides-tutorials/speech-to-text/how-to/keyterms) support for `saaras:v4` on the REST and Batch APIs. Pass a JSON list of up to 50 domain-specific names, places, brands, or technical terms under `keyterms` to bias recognition toward them.