every_holiday/VRCT
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'bugfix_speakerdevice' into develop

Browse Source
This commit is contained in:
misyaguziya
2024-06-20 17:00:14 +09:00
parent 2f3c34ca43 5bb3152d02
commit 9aa4b1447f
1 changed files with 9 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
models/transcription/transcription_transcriber.py
View File

@@ -10,6 +10,7 @@ from .transcription_whisper import getWhisperModel, checkWhisperWeight
import torch import torch
import numpy as np import numpy as np
from pydub import AudioSegment
PHRASE_TIMEOUT = 3 PHRASE_TIMEOUT = 3
MAX_PHRASES = 10 MAX_PHRASES = 10
@@ -104,6 +105,14 @@ class AudioTranscriber:
wf.setframerate(self.audio_sources["sample_rate"]) wf.setframerate(self.audio_sources["sample_rate"])
wf.writeframes(self.audio_sources["last_sample"]) wf.writeframes(self.audio_sources["last_sample"])
temp_file.seek(0) temp_file.seek(0)
if self.audio_sources["channels"] > 2:
audio = AudioSegment.from_file(temp_file, format="wav")
mono_audio = audio.set_channels(1)
temp_file = BytesIO()
mono_audio.export(temp_file, format="wav")
temp_file.seek(0)
with AudioFile(temp_file) as source: with AudioFile(temp_file) as source:
audio = self.audio_recognizer.record(source) audio = self.audio_recognizer.record(source)
return audio return audio