Filipi Fuchter
8577139d21
Fixing to keep the last max samples.
2025-04-17 11:39:06 -03:00
Filipi Fuchter
e2fbbb4b40
Renaming the smart turn classes.
2025-04-17 10:43:21 -03:00
Filipi Fuchter
88ce117e84
Changing the max duration default value to 16 seconds.
2025-04-17 10:35:13 -03:00
Filipi Fuchter
266537c3f4
Fixing to respect the stop_secs.
2025-04-17 10:07:08 -03:00
Filipi Fuchter
230d2f80fa
Merge branch 'main' into smart_turn
2025-04-17 09:36:30 -03:00
Filipi Fuchter
3f0688aefa
Testing smart turn using stop_secs as 5 seconds
2025-04-17 09:36:03 -03:00
Filipi da Silva Fuchter
5be3e6979e
Merge pull request #1533 from pipecat-ai/daily_small_webrtc
...
Example interoping between SmallWebRTC and Daily
2025-04-17 09:19:23 -03:00
Mark Backman
2dafbee2aa
Code review fixes
2025-04-16 22:29:33 -04:00
Mark Backman
1e0a9d7b06
Add previous_text context to ElevenLabsHttpTTSService
2025-04-16 22:22:08 -04:00
Mark Backman
4a23e138b1
Added word/timestamp pairs to ElevenLabsHttpTTSService
2025-04-16 22:20:51 -04:00
Mark Backman
384f80983f
Added word/timestamp pairs to ElevenLabsHttpTTSService
2025-04-16 21:55:00 -04:00
Aleix Conchillo Flaqué
f385cc0460
pyproject: add websockets as google dependency
2025-04-16 15:19:25 -07:00
Aleix Conchillo Flaqué
e97de43de2
add LLMUserAggregatorParams and LLMAssistantAggregatorParams
2025-04-16 15:19:19 -07:00
Aleix Conchillo Flaqué
8299c96ad4
Merge pull request #1603 from pipecat-ai/aleix/deepgram-tavus-fixes
...
deepgram/tavus fixes
2025-04-16 14:55:45 -07:00
Aleix Conchillo Flaqué
e9af585edd
DeepgramTTSService: re-add base_url to constructor
2025-04-16 14:54:02 -07:00
Aleix Conchillo Flaqué
31f7082d12
DeepgramTTSService: use Deepgram's asyncrest instead of asyncio.to_thread
2025-04-16 14:40:59 -07:00
Aleix Conchillo Flaqué
6cea71270e
tts: use smaller audio chunk sizes
2025-04-16 14:40:59 -07:00
Aleix Conchillo Flaqué
d05b2d0e8d
TavusVideoService: fix rate limiting and max size
2025-04-16 14:40:59 -07:00
Filipi Fuchter
5bbf1d0209
Example interoping between SmallWebRTC and Daily.
2025-04-16 17:14:12 -03:00
Mark Backman
235cd9cecc
Merge pull request #1586 from rahultayal22/rah_google_vertex_issue
...
Fixed params issue in Google Vertex ai
2025-04-16 14:56:46 -04:00
Rahul Tayal
ac64f0ba91
Run ruff on code
2025-04-16 23:19:09 +05:30
Mark Backman
ce92dfb5ec
Add eject_at_token_exp to Daily REST helpers, modify default values
2025-04-16 12:26:33 -04:00
Filipi Fuchter
8e36bdbed7
Adding some comments to the code.
2025-04-16 09:11:27 -03:00
Filipi Fuchter
cd8bd7f487
Adding some comments to the code.
2025-04-16 08:58:40 -03:00
Filipi Fuchter
616961b487
Stop removing segments from the end
2025-04-16 08:04:38 -03:00
Filipi Fuchter
650d4d9ee2
Changing the start speech time and adding logs.
2025-04-16 07:55:20 -03:00
Filipi Fuchter
2627cb6bf2
Allowing to define SmartTurnParams
2025-04-16 07:13:13 -03:00
Filipi Fuchter
0e4115049b
Refactoring to use keep alive sessions.
2025-04-16 06:44:57 -03:00
Filipi Fuchter
3ebef9346f
Adding support for RemoteSmartTurn
2025-04-16 06:33:42 -03:00
Filipi Fuchter
3e2d21779f
Refactoring the BaseEndOfTurnAnalyzer to include most of the logic
2025-04-16 06:11:56 -03:00
Filipi Fuchter
cfefcac35f
Resetting the silence frames when the user speaks.
2025-04-15 20:51:36 -03:00
Filipi Fuchter
57b39c084f
Triggering to check if the turn is complete based on the maximum timeout
2025-04-15 20:42:41 -03:00
Filipi Fuchter
11b6de0900
Triggering to check if the turn is complete each time the user stops speaking based on the vad
2025-04-15 17:28:00 -03:00
Mark Backman
ad40a0f076
Update OpenAILLMService and OpenPipeLLMService to use gpt-4.1 by default
2025-04-15 15:11:05 -04:00
Filipi Fuchter
e6325a8229
Integrating with the smart turn model to predict
2025-04-15 16:01:09 -03:00
Filipi Fuchter
73874f6ec0
Loading the smart turn model.
2025-04-15 12:11:06 -03:00
Filipi Fuchter
6ab9a8ad7f
Starting to create a local smart turn
2025-04-15 11:24:39 -03:00
Filipi Fuchter
821e303249
Bringing Aleix initial implementation for the smart turn.
2025-04-15 10:21:40 -03:00
chadbailey59
efae26a5a8
Client connect/disconnect events for DailyTransport ( #1544 )
...
* added multi transport example
* added working example
* restructured example and added readme
* removed image
* cleanup
* changed data type of callback signature
* removed pipecat example
* added changelog
2025-04-14 15:56:41 -05:00
Rahul Tayal
001c26b79c
Fixed params issue in Google Vertex ai
2025-04-14 23:29:16 +05:30
Aleix Conchillo Flaqué
ab6be11a0e
SoundfileMixer: add mixing argument and require keywords
2025-04-14 08:30:56 -07:00
Filipi Fuchter
e1108466f6
Fixed an issue in SmallWebRTCTransport where an error was thrown if the client did not create a video transceiver.
2025-04-14 10:36:25 -03:00
Mark Backman
f6accbd510
Updating foundation examples to use SmallWebRTCTransport and pipecat-ai-small-webrtc-prebuilt ( #1534 )
...
Co-authored-by: Filipi Fuchter <filipi@daily.co >
2025-04-11 19:44:16 -04:00
Mark Backman
b9a2ed5b58
Fix: GeminiMultimodalLiveLLMService, add spaces between words in assistant context messages
2025-04-11 15:14:52 -04:00
Mark Backman
7ac12ffc85
Merge pull request #1550 from pipecat-ai/mb/cartesia-spelling-timestamps
...
Fix: Cartesia's spelling feature adds whole word to context
2025-04-11 15:13:55 -04:00
Filipi da Silva Fuchter
f623cf96f7
Merge pull request #1560 from pipecat-ai/bot_left_signalling
...
Bot left signalling message
2025-04-11 16:08:01 -03:00
Mark Backman
06be20eb16
Fix: Cartesia's spelling feature adds whole word to context
2025-04-11 15:04:58 -04:00
Filipi Fuchter
816b3a9545
Fixing ruff format
2025-04-11 15:37:16 -03:00
Filipi Fuchter
255666925b
Sending a new signalling message peerLeft.
2025-04-11 15:35:50 -03:00
Mark Backman
e28c199dd1
Add GeminiMultimodalLiveLLMService support for VAD Params
2025-04-11 11:46:48 -04:00
