From 24367d459b6487066c75d2cfb6658fa4f203c28f Mon Sep 17 00:00:00 2001 From: Tom <29130704+TomBayne@users.noreply.github.com> Date: Sun, 13 Apr 2025 13:03:57 +0100 Subject: [PATCH] Enable vad_filter to improve quality of transcription in faster-whisper model. --- backend/open_webui/routers/audio.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/backend/open_webui/routers/audio.py b/backend/open_webui/routers/audio.py index a5447e1fc..c5d5f99af 100644 --- a/backend/open_webui/routers/audio.py +++ b/backend/open_webui/routers/audio.py @@ -497,7 +497,7 @@ def transcribe(request: Request, file_path): ) model = request.app.state.faster_whisper_model - segments, info = model.transcribe(file_path, beam_size=5) + segments, info = model.transcribe(file_path, beam_size=5, vad_filter=True) log.info( "Detected language '%s' with probability %f" % (info.language, info.language_probability)