enh: WHISPER_MULTILINGUAL

backend/open_webui/config.py

@@ -3744,11 +3744,9 @@ WHISPER_MODEL_AUTO_UPDATE = (
     and os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true"
 )
 
-WHISPER_VAD_FILTER = PersistentConfig(
-    "WHISPER_VAD_FILTER",
-    "audio.stt.whisper_vad_filter",
-    os.getenv("WHISPER_VAD_FILTER", "False").lower() == "true",
-)
+WHISPER_VAD_FILTER = os.getenv("WHISPER_VAD_FILTER", "False").lower() == "true"
+
+WHISPER_MULTILINGUAL = os.getenv("WHISPER_MULTILINGUAL", "False").lower() == "true"
 
 WHISPER_LANGUAGE = os.getenv("WHISPER_LANGUAGE", "").lower() or None
 

backend/open_webui/main.py

@@ -1174,7 +1174,6 @@ app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL
 app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY
 
 app.state.config.WHISPER_MODEL = WHISPER_MODEL
-app.state.config.WHISPER_VAD_FILTER = WHISPER_VAD_FILTER
 app.state.config.DEEPGRAM_API_KEY = DEEPGRAM_API_KEY
 
 app.state.config.AUDIO_STT_AZURE_API_KEY = AUDIO_STT_AZURE_API_KEY

backend/open_webui/routers/audio.py

@@ -41,8 +41,10 @@ from open_webui.config import (
     WHISPER_MODEL_AUTO_UPDATE,
     WHISPER_COMPUTE_TYPE,
     WHISPER_MODEL_DIR,
+    WHISPER_VAD_FILTER,
     CACHE_DIR,
     WHISPER_LANGUAGE,
+    WHISPER_MULTILINGUAL,
     ELEVENLABS_API_BASE_URL,
 )
 
@@ -601,8 +603,9 @@ def transcription_handler(request, file_path, metadata, user=None):
         segments, info = model.transcribe(
             file_path,
             beam_size=5,
-            vad_filter=request.app.state.config.WHISPER_VAD_FILTER,
+            vad_filter=WHISPER_VAD_FILTER,
             language=languages[0],
+            multilingual=WHISPER_MULTILINGUAL,
         )
         log.info(
             "Detected language '%s' with probability %f"