> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://developers.deepgram.com/voice-agent/optimize/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://developers.deepgram.com/_mcp/server.

# Optimize

## Docs

- [Voice Agent TTS Controls](https://developers.deepgram.com/docs/voice-agent-tts-controls.md): Apply speed, expressivity, pronunciation, and pacing controls inside Voice Agent sessions.
- [Voice Agent Message Flow](https://developers.deepgram.com/docs/voice-agent-message-flow.md): Implement the correct WebSocket message sequence for Voice Agent conversations.
- [Speculative Replies & Turn Confirmation](https://developers.deepgram.com/docs/voice-agent-speculative-replies.md): The agent starts building a reply before it is certain the user has finished speaking. Audio waits for confirmation. Function calls do not, unless you ask them to.
- [Session Observability](https://developers.deepgram.com/docs/voice-agent-observability.md): Log client and server WebSocket messages for per-session, turn-by-turn observability into transcripts, agent behavior, function calls, errors, and latency.
- [Voice Agent Audio & Playback](https://developers.deepgram.com/docs/voice-agent-audio-playback.md): Learn some strategies for dealing with audio and playback issues.
- [Audio Preprocessing & Barge-In](https://developers.deepgram.com/voice-agent/optimize/audio-preprocessing-barge-in.md): When noise suppression and echo cancellation help speech-to-text — and when they hurt.
- [Voice Agent Adaptive Echo Cancellation](https://developers.deepgram.com/docs/voice-agent-echo-cancellation.md): Learn some tips and strategies for dealing with echo cancellation.