Mark Backman
a9b1298f3b
Fix: RTVIObserver now outputs a single bot started and stopped speaking event per turn
2025-03-24 10:25:31 -04:00
Filipi Fuchter
3ed764a769
Only checking the length if tools is a list.
2025-03-21 12:56:05 -03:00
Aleix Conchillo Flaqué
442f18d47b
ultravox: fix formatting
2025-03-20 19:19:08 -07:00
Aleix Conchillo Flaqué
fc78e6fc5a
ElevenLabs: add support for a sample rate of 8000
2025-03-20 19:13:23 -07:00
milo157
3b4d91e1c1
Fixed ultravox service bugs ( #1420 )
2025-03-20 18:55:43 -07:00
Mark Backman
2ac8f2ec2d
Fix linting
2025-03-20 18:40:16 -04:00
Mark Backman
541a4b6063
Update InputAudioTranscription to use gpt-4o-transcribe model, update 19 examples to use FunctionSchema
2025-03-20 18:23:24 -04:00
Aleix Conchillo Flaqué
96fa6c19a8
Merge pull request #1398 from nicougou/feature/openai_custom_headers
...
feature: add custom headers to AsyncOpenAI
2025-03-20 13:45:57 -07:00
Aleix Conchillo Flaqué
08956e914a
livekit: remove unnecessary transport cleanup() function
2025-03-20 11:23:40 -07:00
Aleix Conchillo Flaqué
5a39f146f6
LLMUserContextAggregator: fix emulated user started/stopped speaking issues
2025-03-20 11:23:40 -07:00
Aleix Conchillo Flaqué
efa5f133d7
openai_realtime: fix and update function calling
2025-03-20 11:14:59 -07:00
Paul Kompfner
721ee75887
Comment tweak
2025-03-20 13:43:00 -04:00
Paul Kompfner
ada68f0699
More robust handling of conversation item retrieval errors in OpenAIRealtimeBetaLLMService
2025-03-20 13:43:00 -04:00
Mark Backman
70dbf0d6fc
Updated default models for OpenAISTTService and OpenAITTSService to gpt-4o based models
2025-03-20 13:42:56 -04:00
Paul Kompfner
f0774268cc
Rename gpt-4o-transcribe-latest to gpt-4o-transcribe in OpenAIRealtimeBetaLLMService
2025-03-20 13:39:40 -04:00
Paul Kompfner
f94a099111
Revert the default model to be "gpt-4o-realtime-preview-2024-12-17" In OpenAIRealtimeBetaLLMService
2025-03-20 13:39:36 -04:00
Paul Kompfner
7b594093dd
Handle the possibility of multiple concurrent calls to retrieve_conversation_item() in the OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
Paul Kompfner
31317ce77d
Add error handling to the retrieve_conversation_item() method of the OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
Paul Kompfner
f693a3c70f
Add retrieve_conversation_item() method to OpenAIRealtimeBetaLLMService, using the new conversation.item.retrieve introspection message.
2025-03-20 13:31:28 -04:00
Paul Kompfner
39ca607bbb
Add on_conversation_item_created and on_conversation_item_updated events to OpenAIRealtimeBetaLLMService.
...
The hope is that this will expose to the user conversation item ids at relevant times for them to use with the new `conversation.item.retrieve` introspection message.
2025-03-20 13:31:28 -04:00
Paul Kompfner
9840abd85b
Make it so you specifying model=None when creating a InputAudioTranscription results in a validation error
2025-03-20 13:31:28 -04:00
Paul Kompfner
1075c25055
Add new semantic turn detection option to OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
Paul Kompfner
e91610c69e
linter fix
2025-03-20 13:31:28 -04:00
Paul Kompfner
1a20d9bed7
Add new input_audio_noise_reduction option to OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
Paul Kompfner
d009b80438
Add new GPT-4o transcription option to OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
kompfner
fe5fc30211
Revert "Add new GPT-4o transcription option to OpenAIRealtimeBetaLLMService"
2025-03-20 13:31:28 -04:00
Paul Kompfner
be2cf6d556
formatting fix
2025-03-20 13:31:28 -04:00
Paul Kompfner
e80bfe22de
Add new GPT-4o transcription option to OpenAIRealtimeBetaLLMService
2025-03-20 13:31:28 -04:00
Paul Kompfner
214c8f79eb
linter fix
2025-03-20 13:31:28 -04:00
Paul Kompfner
16accafa6d
formatting fix
2025-03-20 13:31:28 -04:00
Kwindla Hultman Kramer
4449e9a25b
add response.done status=failed error
2025-03-20 13:31:28 -04:00
Kwindla Hultman Kramer
2b4debec11
add support for conversation.item.input_audio_transcription.delta
2025-03-20 13:31:28 -04:00
Mark Backman
f298febacf
Add FalSTTService
2025-03-20 12:45:16 -04:00
Aleix Conchillo Flaqué
e0c3f6ad83
services: mark function calls as completed even the result is None
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
b1d506c137
GoogleAssistantContextAggregator: properly update function response
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
1f6ed01ba6
LLMAssistantContextAggregator: remove tool call id with image requests
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
3e9678db84
user image requests can now be related to function calls
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
d455fd070e
update CHANGELOG
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
d1550d5a85
tests: remove TestFrameProcessor, reimplement with run_test()
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
a98000fd1d
function calling now run in tasks
2025-03-20 08:51:25 -07:00
Aleix Conchillo Flaqué
fc06306efd
Merge pull request #1406 from pipecat-ai/aleix/pipeline-task-idle-timeouts
...
PipelineTask: automatically cancel tasks if pipeline is idle
2025-03-20 08:37:39 -07:00
Aleix Conchillo Flaqué
2417ec4f92
LLMUserContextAggregator: increase bot_interruption_timeout to 5 seconds
2025-03-20 01:20:34 -07:00
Aleix Conchillo Flaqué
7cdcd1c3d1
OpenAITTSService: allow specifying any model name
2025-03-20 01:20:34 -07:00
Aleix Conchillo Flaqué
b6be25ab84
SegmentedSTTService: use VAD events to detect valid audio
2025-03-20 00:31:49 -07:00
Aleix Conchillo Flaqué
e18d9f6a11
PipelineTask: automatically cancel tasks if pipeline is idle
2025-03-19 23:30:46 -07:00
Aleix Conchillo Flaqué
8d58d1c8bb
Merge pull request #1404 from pipecat-ai/aleix/gemini-push-frame-fixes
...
GeminiMultimodalLiveLLMService: fix duplicated messages in context
2025-03-19 21:51:39 -07:00
Aleix Conchillo Flaqué
a3b5e4413a
WebsocketTTSService: add on_connection_error and reconnect_on_error
2025-03-19 19:38:08 -07:00
Aleix Conchillo Flaqué
8942c2e053
GeminiMultimodalLiveLLMService: fix duplicated messages in context
...
Fixes #1384
2025-03-19 15:33:54 -07:00
Aleix Conchillo Flaqué
48d73a2636
SegmentedSTTService: allow audio to pass-through downstream
2025-03-19 11:06:12 -07:00
Aleix Conchillo Flaqué
336e2f1579
TTSServices: for now just specify a single text aggregator
2025-03-19 11:02:29 -07:00
