Aleix Conchillo Flaqué
a9e565f355
processors: fix input queue interruptions
2024-11-06 13:12:24 -08:00
Aleix Conchillo Flaqué
49005d02f5
services(tts): use TTSSpeakFrame in say() method
2024-11-06 13:05:03 -08:00
Aleix Conchillo Flaqué
6d8b885071
transports(base_output): push bot started/stopped frames downstream
2024-11-06 13:04:37 -08:00
Aleix Conchillo Flaqué
2eccb33e73
processors: allow passing a callback when queued frame is processed
2024-11-06 13:04:37 -08:00
Aleix Conchillo Flaqué
22ca4c5a02
processors: cancel input task and empty queue with interruptions
2024-11-06 13:04:37 -08:00
Aleix Conchillo Flaqué
84f26ac1ca
processors: introduce input queues
...
Frame processors can now decide if they should continue processing frames or
not, and if so also decide when to continue processing frames. For example,
asynchronous TTS services will stop processing frames until they have generated
all the audio for an LLM response.
2024-11-06 12:13:49 -08:00
Aleix Conchillo Flaqué
bd50201ce4
transports(daily): just make it clear we subscribe to camera
2024-11-04 17:32:46 -08:00
Aleix Conchillo Flaqué
358c458265
transports(base_input): handle filter contorl frames
2024-11-04 16:19:52 -08:00
Aleix Conchillo Flaqué
807dbbe326
audio(noisereduce): allow enabling/disabling filter
2024-11-04 16:13:29 -08:00
Aleix Conchillo Flaqué
3c116b291d
audio(mixers): some cosmetics
2024-11-04 15:37:08 -08:00
Aleix Conchillo Flaqué
0dd413ee90
audio(filters): add noisereduce filter
2024-11-04 15:37:08 -08:00
Aleix Conchillo Flaqué
abc8ede3d7
introduce audio filters
2024-11-04 15:37:08 -08:00
Aleix Conchillo Flaqué
126324ca1b
Merge pull request #687 from pipecat-ai/aleix/transport-audio-mixers
...
introduce transport audio mixers
2024-11-04 13:14:36 -08:00
Aleix Conchillo Flaqué
0ac9e2dd3f
transports(network): synchronize with time before sending data
2024-11-04 13:04:18 -08:00
Aleix Conchillo Flaqué
81c476dd4c
introduce output transport audio mixers
2024-11-03 11:13:02 -08:00
Kwindla Hultman Kramer
151242d3a0
Merge pull request #666 from pipecat-ai/khk/realtime-pipecat-vad
...
Support using Pipecat turn detection instead of OpenAI Realtime API turn detection
2024-11-02 08:36:31 -07:00
Kwindla Hultman Kramer
93c6e5098c
added comment explaining config of TurnDetection
2024-11-02 08:24:54 -07:00
Aleix Conchillo Flaqué
4455b2a428
rtvi: create queues before tasks
2024-11-01 23:06:50 -07:00
Aleix Conchillo Flaqué
94062592ef
base_output: generate smaller audio frames of the same incoming type
2024-11-01 23:06:50 -07:00
Aleix Conchillo Flaqué
d2401a76c8
base_output: only generate bot speaking with TTS audio frames
2024-11-01 23:06:50 -07:00
Mark Backman
802c29e9e1
Add recording events and callbacks
2024-11-01 10:20:00 -04:00
Aleix Conchillo Flaqué
018e51e8a3
add notifiers and more frame filters
2024-10-30 16:36:17 -07:00
Vanessa Pyne
b050143952
Merge pull request #676 from RonakAgarwalVani/fix/chunk-choices-delta-none
...
Fix uncaught exception when accessing 'tool_calls' in NoneType delta in response handling
2024-10-30 14:44:32 -05:00
Mark Backman
98ea1f0791
Merge pull request #675 from pipecat-ai/mb/playht-add-request-id
...
Add a request_id to each TTS sequence
2024-10-30 13:56:15 -04:00
Mark Backman
8272c35527
Use a request_id in TTS commands for the PlayHT websocket service
2024-10-30 13:54:18 -04:00
Mark Backman
e973e82e05
Merge pull request #672 from pipecat-ai/mb/fix-playht
...
Fix PlayHT TTFB metrics
2024-10-30 13:53:02 -04:00
RonakAgarwalVani
d1396bf618
Update openai.py
2024-10-30 14:26:49 +05:30
Mark Backman
2e57d21b89
Fix ttfb metrics
2024-10-28 20:27:24 -04:00
Aleix Conchillo Flaqué
e3ce619284
transports(base_output): use audio frames for bot speaking detection
2024-10-28 10:07:37 -07:00
Aleix Conchillo Flaqué
b32448e967
transports(daily): make functions async
2024-10-25 15:01:52 -07:00
Aleix Conchillo Flaqué
b905b57dfc
transports(daily): allow updating subscriptions before join
2024-10-25 14:46:17 -07:00
Kwindla Hultman Kramer
e1a7edfb58
make it possible to use Pipecat turn detection instead of OpenAI turn detection
2024-10-25 15:59:48 -05:00
Aleix Conchillo Flaqué
55026898f6
transports(base_output): use vad stop secs for bot stopped speaking
2024-10-25 12:59:15 -07:00
Aleix Conchillo Flaqué
4283557894
audio(vad): expose params property
2024-10-25 12:59:15 -07:00
Aleix Conchillo Flaqué
5ab00e01aa
transports(base_output): fix constant bot started/stopped speaking frames
2024-10-25 12:10:24 -07:00
Aleix Conchillo Flaqué
4eacb34fd8
services(aws): read stream and resample in a thread
2024-10-25 11:22:28 -07:00
Aleix Conchillo Flaqué
3a8aacccf7
Merge pull request #663 from pipecat-ai/aleix/audio-resampling-with-resampy
...
audio: use resamply for audio resampling
2024-10-25 10:16:20 -07:00
roey
54c0bf0c70
Adding TavusVideoService layer ( #617 )
...
Co-authored-by: roey <159067767+roey-tavus@users.noreply.github.com >
Co-authored-by: Mert Gerdan <mert@tavus.io >
Co-authored-by: Aleix Conchillo Flaqué <aleix@daily.co >
2024-10-25 09:46:25 -07:00
Aleix Conchillo Flaqué
778b05a252
audio: use resamply for audio resampling
2024-10-25 09:22:22 -07:00
Mark Backman
f16a416c2b
Merge pull request #660 from pipecat-ai/mb/add-gemini-inputs
...
Add input params to Google Gemini
2024-10-24 20:58:19 -04:00
Mark Backman
37820ac0df
Add input params to Google Gemini
2024-10-24 20:12:41 -04:00
Aleix Conchillo Flaqué
8ea80d43f4
transports(daily): only transcribe user audio
2024-10-24 17:06:43 -07:00
Aleix Conchillo Flaqué
cfb48200c2
services(azure): support sample rates
2024-10-24 16:47:35 -07:00
Aleix Conchillo Flaqué
6d317c6e8e
audio: don't resample if same sample rate
2024-10-24 16:47:35 -07:00
Aleix Conchillo Flaqué
158d52856f
transports(livekit): fix VADAnalyzer import
2024-10-24 16:47:35 -07:00
Aleix Conchillo Flaqué
92a69e404f
update TTS and transport output sample rate to 24000
2024-10-24 16:47:35 -07:00
Aleix Conchillo Flaqué
d24c6185d8
Merge pull request #654 from pipecat-ai/aleix/daily-allow-completion-futures
...
transport(daily): allow completion futures
2024-10-24 14:28:53 -07:00
Mark Backman
1fd21578a6
Merge pull request #657 from pipecat-ai/mb/add-elevenlabs-output-format-type
...
Add ElevenLabs output format type
2024-10-24 17:07:04 -04:00
Mark Backman
700db87127
Merge pull request #656 from pipecat-ai/mb/add-gemini-metrics
...
Add Gemini token usage metrics
2024-10-24 17:04:56 -04:00
Mark Backman
6f1310569c
Add ElevenLabs output format type
2024-10-24 17:03:45 -04:00
