v0.55.0 v0.55.1
low risk
maurosoria/dirsearchOct 2, 2026 → Oct 2, 20260 days between releases
What's New
Default TTS switched to Deepgram Aura and pause-mode barge-in audio dropping is fixed.
Improvements
- Default TTS provider changed to Deepgram Aura for faster time to first audio.
- OpenAI TTS configuration defaults updated to `tts-1` for better stability.
- Improved TTS performance metrics and reduced latency in synthesis.
Bug Fixes
- Fixed pause-mode barge-in dropping bot audio after resolution.
- Resolved audio backlog issues during pause-mode arbitration.
Upgrade Notes
The default Text-to-Speech provider has changed to Deepgram Aura. If you rely on OpenAI, set `BOT_TTS_PROVIDER=openai` and `BOT_TTS_MODEL=tts-1`. No code changes are required.
The Two Releases
v0.55.0
v0.55.0
Oct 2, 2026· 20 days ago
SiphonAI v0.55.0 introduces a new Pipecat transport for real-time audio pacing and updates security dependencies.
Full release notesv0.55.1
v0.55.1
Oct 2, 2026· 20 days ago
SiphonAI v0.55.1 improves text-to-speech performance by defaulting to Deepgram Aura and fixes audio dropping issues in pause-mode barge-ins.
Full release notes