Update app.py

app.py

@@ -25,10 +25,10 @@ dict_languages = {"English": "en",
                   "Hindi": "hi"}
 
 @spaces.GPU
-def process_transcript(audio_path, model, processor, language):
+def process_transcript(model, processor, audio_path, language):
     """Process audio with selected Voxtral model and return the generated response"""
-    
-    inputs = processor.apply_transcrition_request(language=language, audio=audio_path, model_id=model_name)
+    id_language = dict_languages[language]
+    inputs = processor.apply_transcrition_request(language=id_language, audio=audio_path, model_id=model_name)
     inputs = inputs.to(device, dtype=torch.bfloat16)
 
     outputs = model.generate(**inputs, max_new_tokens=MAX_TOKENS)
@@ -62,8 +62,8 @@ with gr.Blocks(title="Transcription") as transcript:
 
 
     submit_transcript.click(
-        fn=process_transcript,
-        inputs=[dict_languages[sel_language], model, processor, sel_audio],
+        ,fn=lambda txt: process_transcript(model, processor, sel_audio, sel_language)
+        inputs=[sel_language, sel_audio],
         outputs=text_transcript
     )