0.0.20

What's Changed

• Improve error message if track kind and modality mismatch by @marcusvaltonen in #230
• Ignore output_frame_size parameter by @vvolhejn in #210
• Improve error handling for websockets by @vvolhejn in #238
• Add track_constraints to Stream class by @freddyaboulton in #241
• Support CloseStream in Websocket by @freddyaboulton in #242
• MIT license by @freddyaboulton in #246
• Allow extra tracks (#231) by @marcusvaltonen in #249
• Add ability to trigger ReplyOnPause without waiting for pause by @freddyaboulton in #250
• Fix bug where additional outputs where not fetched immediately on subsequent connections by @freddyaboulton in #254
• Introduce automatic linting with Github workflows by @marcusvaltonen in #251
• Add API Reference and llms.txt by @freddyaboulton in #256
• Dont run docs ci on prs from forks by @freddyaboulton in #257
• Introduce static type checking with pyright by @marcusvaltonen in #255
• Introduce unit tests by @marcusvaltonen in #248
• Add started_talking log message in ReplyOnPause and in api.md by @erwaen in #260
• Enforce modern typing by @marcusvaltonen in #258
• Cloudflare turn integration by @freddyaboulton in #264
• 0.0.20 release by @freddyaboulton in #265

New Contributors

• @erwaen made their first contribution in #260

Full Changelog: 0.0.19...0.0.20

0.0.19

What's Changed

• Add text-to-speech-gallery + reword galleries to be "Plugin Ecosystem" by @freddyaboulton in #218
• Remove twice instantiated event by @marcusvaltonen in #221
• Add Kroko-ASR model to STT gallery by @sgarg26 in #219
• Close Stream from Backend by @freddyaboulton in #222
• Add get_context function that can be used to retrieve the webrtc id by @freddyaboulton in #223
• Trigger 0.0.19 release by @freddyaboulton in #226

New Contributors

• @marcusvaltonen made their first contribution in #221
• @sgarg26 made their first contribution in #219

Full Changelog: 0.0.18...0.0.19

0.0.18

What's Changed

• Fix gradio concurrency issue in fetching additional outputs by @freddyaboulton in #211
• Trigger 0.0.18 by @freddyaboulton in #212

gradio-concurrency-fixed.mp4

Full Changelog: 0.0.17...0.0.18

0.0.17

What's Changed

• Add js assets by @freddyaboulton in #192
• Create py.typed by @vvolhejn in #196
• Added HumAwareVAD to VAD Gallery by @CuriousMonkey7 in #194
• Minor formating change to the turn gallery by @freddyaboulton in #197
• Some Video Fixes by @freddyaboulton in #200
• Add support for trickle ice by @freddyaboulton in #193
• trigger release by @freddyaboulton in #201

Full Changelog: 0.0.16...0.0.17

0.0.16

What's Changed

• Rename the Vad Gallery to Turn Taking Gallery by @freddyaboulton in #173
• add fastrc with Elecron app example to cookbook by @swairshah in #178
• Add on-device whisper example to cookbook by @sofi444 in #179
• Add example for "Talk to Azure OpenAi" by @MLYengineering in #181
• Fix Fastphone bug with latest gradio version by @freddyaboulton in #183
• Fix outdated import in docs by @vvolhejn in #185
• Fix issue when the audio stream mixes sample rates and numpy array data types by @freddyaboulton in #188
• Add path argument to Stream.mount to let developers mount multiple streams in the same app easily by @freddyaboulton in #189
• Bump version to 0.0.16 by @freddyaboulton in #190

New Contributors

• @swairshah made their first contribution in #178
• @sofi444 made their first contribution in #179
• @MLYengineering made their first contribution in #181
• @vvolhejn made their first contribution in #185

Full Changelog: 0.0.15...0.0.16

0.0.15

What's Changed

• feat: Add optional startup function to ReplyOnPause by @Ryu1845 in #170

We've added a new parameter to ReplyOnPause and ReplyOnStopWords called startup_fn. Pass in generator that yields any output data and the assistant will run that generator as soon as a user connects.

More information here: https://fastrtc.org/userguide/audio/#startup-function

Audio Response

welcome_message.mp4

Text Response

sambanova_welcome.mp4

New Contributors

• @Ryu1845 made their first contribution in #170

Full Changelog: 0.0.14...0.0.15

0.0.14

Main Features

Microphone Muting

Ability to load community VAD and Text-to-Speech Models

What's Changed

• Adding nextjs + 11labs + openai streaming demo by @rohanprichard in #139
• Add Method for loading community Vad Models by @freddyaboulton in #136
• Community STT models by @freddyaboulton in #147
• Community stt models by @freddyaboulton in #149
• fix: unused user-provided Silero option by @CuriousMonkey7 in #150
• Raise WebRTC Errors in "Receive" case by @freddyaboulton in #151
• Raise error if code in any part of video processing fails by @freddyaboulton in #153
• fix: ensure 'model' is copied in ReplyOnPause.copy() by @CuriousMonkey7 in #155
• Fix Warning in Advanced Configuration by @freddyaboulton in #157
• Add microphone mute by @freddyaboulton in #158
• Added to gallery by @Codeblockz in #159
• Copy Model on ReplyOnStopWords by @freddyaboulton in #160
• Add docs on how to contribute by @freddyaboulton in #161
• Release 0.0.14 by @freddyaboulton in #164

New Contributors

• @rohanprichard made their first contribution in #139
• @Codeblockz made their first contribution in #159

Full Changelog: 0.0.13...0.0.14

0.0.13

What's Changed

• Fix kokoro batch issue by @freddyaboulton in #128
• Fix kokoro batch issue by @freddyaboulton in #133
• UnboundLocalError: local variable 'button' referenced before assignment by @akjava in #126
• Improve Interruption Handling by @freddyaboulton in #134
• feat: Added documentation for twilio integration by @mahimairaja in #125
• Simplify Cloudflare config with new endpoint by @mhart in #135
• Add subtitle to UIArgs by @freddyaboulton in #137
• Some video send-receive bug fixes by @freddyaboulton in #145
• Release 0.0.13 by @freddyaboulton in #146

New Contributors

• @akjava made their first contribution in #126
• @mahimairaja made their first contribution in #125

Full Changelog: 0.0.11...0.0.13

0.0.12

What's Changed

• Clean up interrupted generators in ReplyOnPause/ReplyOnStopWords by @freddyaboulton in #124
• Link + doc fixes

Full Changelog: 0.011...0.0.12

Version 0.0.11

• ReplyOnPause and ReplyOnStopWords can now be interrupted by default. This means that if the user starts to speak while the generator is running, the generator will be interrupted and the new response will start.

Also included some minor fixes to some transcription demos.

2025-01-27.16-07-29.mov