wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

services(whisper): increase no speech probability to 0.4

Browse Source
This commit is contained in:
Aleix Conchillo Flaqué
2024-06-06 23:39:44 -07:00
parent 90d11398e6
commit b4340d0185
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/pipecat/services/whisper.py
View File

@@ -45,7 +45,7 @@ class WhisperSTTService(STTService):
model: Model = Model.DISTIL_MEDIUM_EN, model: Model = Model.DISTIL_MEDIUM_EN,
device: str = "auto", device: str = "auto",
compute_type: str = "default", compute_type: str = "default",
no_speech_prob: float = 0.1, no_speech_prob: float = 0.4,
**kwargs): **kwargs):
super().__init__(**kwargs) super().__init__(**kwargs)

2
src/pipecat/transports/base_input.py
View File

@@ -177,7 +177,7 @@ class BaseInputTransport(FrameProcessor):
vad_state = self._handle_vad(frame.audio, vad_state) vad_state = self._handle_vad(frame.audio, vad_state)
audio_passthrough = self._params.vad_audio_passthrough audio_passthrough = self._params.vad_audio_passthrough
# Push audio downstream if passthrough. # Push audio downstream if passthrough.
if audio_passthrough: if audio_passthrough:
future = asyncio.run_coroutine_threadsafe( future = asyncio.run_coroutine_threadsafe(
self._internal_push_frame(frame), self._loop) self._internal_push_frame(frame), self._loop)