Update Pipecat integration for Waves v4.0.0 API#51
Conversation
- Remove "not yet merged" status — integration is now live - Update model names to lightning_v3.1 / lightning_v3.1_pro (underscore) - Document default model (lightning_v3.1_pro) and per-model default voices - Add output_format setting reference table - Fix imports and constructor API (Settings pattern) - Replace temp branch install with pip install pipecat-ai[smallest]
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
EntelligenceAI PR SummaryUpdates
Confidence Score: 5/5 - Safe to MergeSafe to merge — this PR is a documentation-only update to Key Findings:
Files requiring special attention
|
output_format is a SmallestTTSService constructor param, not a Settings field. Move it to the code example and split the reference table into constructor params and Settings sections.
… to True Mark's implementation moved word_timestamps out of SmallestTTSSettings into __init__ and flipped the default to True (opt-out). Update the reference table to match.
Summary
Updates
integrations/pipecat/README.mdto match the Waves v4.0.0 API changes shipped in pipecat-ai/pipecat#4535.pipecat-ai[smallest]pip install "pipecat-ai[smallest]"lightning_v3.1,lightning_v3.1_prolightning_v3.1_pro(voice:meher);lightning_v3.1defaults tosophiaoutput_formatsetting (pcm,mp3,wav,ulaw,alaw)Settingspattern