> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.vapi.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.vapi.ai/_mcp/server.

# What's New: Week of November 2, 2026

## New Voices

1. **GPT-Live voices**: Nine new [GPT-Live](/gpt-live/configuration) voices are available: Nuri, Juni, and Flitz (German), Brise and Sillage (French), Nira (Hindi), Harema and Shida (Japanese), and Noeul (Korean). Assistants that don't use a GPT-Live model can no longer be saved with a GPT-Live only voice.

## Features

1. **GPT-Live prompt editor**: The [GPT-Live](/gpt-live/configuration) speaker and reasoner prompts now use the rich text editor, with visual and code views and a fullscreen mode.
2. **Reasoning effort for structured outputs**: You can now set a reasoning effort for [structured outputs](/assistants/structured-outputs#reasoning-effort) that use an OpenAI reasoning model.
3. **Deepgram feedback filter setting**: Set `feedbackFilterEnabled` to false on a [Deepgram](/providers/transcriber/deepgram#feedback-filter) transcriber to keep caller speech that repeats what the assistant just said. It's on by default.
4. **Simulation suite generation API**: You can now draft requirements and generate [simulation suites through the API](/observability/simulations-generate-api) with a private API key.

## Fixes and Improvements

1. **ElevenLabs pronunciation dictionaries**: Calls that use an [ElevenLabs](/providers/voice/elevenlabs) pronunciation dictionary without a pinned version no longer fail. They use the dictionary's latest version.
2. **ElevenLabs silent turns**: When ElevenLabs returns no audio for a turn, the call retries once and then switches to your [fallback voice](/voice-fallback-plan).
3. **GPT-Live phone audio**: [GPT-Live](/gpt-live/configuration) phone calls no longer have short gaps of silence in the middle of words.