v0.55.0 v0.55.1
mizorewww/laya-mlxOct 2, 2026 → Oct 2, 20260 days between releases
What's New
Switched default TTS to Deepgram Aura for faster response times
New Features
- Deepgram Aura TTS now defaults in the pipecat-bot-py example
- Added BOT_TTS_MODEL setting for OpenAI TTS configuration
Improvements
- Median time to first audio reduced from 0.94s to 0.52s
- PSTN call gap reduced from 1.95s to 1.01s
Bug Fixes
- Pause-mode barge-in no longer drops bot audio after resolution
Upgrade Notes
If you rely on the OpenAI TTS default, update environment variables to BOT_TTS_PROVIDER=openai and BOT_TTS_MODEL=tts-1. The pause-mode barge-in behavior is now fixed automatically.
The Two Releases
v0.55.0
Oct 2, 2026· 20 days ago
This release introduces a new Pipecat transport that allows answering SiphonAI calls with a Pipecat pipeline, featuring real-time audio pacing. It includes an updated example bot and documentation improvements, along with security updates for the sightglass project.
Full release notesv0.55.1
Oct 2, 2026· 20 days ago
This release optimizes the pipecat-bot-py example by switching the default TTS provider to Deepgram Aura, which offers significantly faster response times. It also fixes a pause-mode barge-in issue where bot audio was being dropped after resolution.
Full release notes