Update app.py

app.py

@@ -191,6 +191,9 @@ def transcribe_audio(input_source, pipeline_type, model_id, dtype, batch_size, d
         logging.info(f"Transcription parameters: pipeline_type={pipeline_type}, model_id={model_id}, dtype={dtype}, batch_size={batch_size}, download_method={download_method}")
         verbose_messages = f"Starting transcription with parameters:\nPipeline Type: {pipeline_type}\nModel ID: {model_id}\nData Type: {dtype}\nBatch Size: {batch_size}\nDownload Method: {download_method}\n"
 
+        if verbose:
+            yield verbose_messages, "", None
+
         if pipeline_type == "faster-batched":
             model = WhisperModel(model_id, device="auto", compute_type=dtype)
             pipeline = BatchedInferencePipeline(model=model)
@@ -221,7 +224,10 @@ def transcribe_audio(input_source, pipeline_type, model_id, dtype, batch_size, d
         if isinstance(input_source, str) and (input_source.startswith('http://') or input_source.startswith('https://')):
             audio_path = download_audio(input_source, download_method)
             verbose_messages += f"Audio file downloaded: {audio_path}\n"
-            if audio_path.startswith("Error"):
+            if verbose:
+                yield verbose_messages, "", None
+
+            if not audio_path or audio_path.startswith("Error"):
                 yield f"Error: {audio_path}", "", None
                 return
         else:
@@ -231,6 +237,8 @@ def transcribe_audio(input_source, pipeline_type, model_id, dtype, batch_size, d
             trimmed_audio_path = trim_audio(audio_path, start_time or 0, end_time)
             audio_path = trimmed_audio_path
             verbose_messages += f"Audio trimmed from {start_time} to {end_time}\n"
+            if verbose:
+                yield verbose_messages, "", None
 
         start_time_perf = time.time()
         if pipeline_type in ["faster-batched", "faster-sequenced"]:
@@ -285,8 +293,13 @@ def transcribe_audio(input_source, pipeline_type, model_id, dtype, batch_size, d
 def update_model_dropdown(pipeline_type):
     model_choices = get_model_options(pipeline_type)
     logging.info(f"Model choices for {pipeline_type}: {model_choices}")
-    return gr.Dropdown.update(choices=model_choices, value=model_choices[0] if model_choices else None)
     
+    # Check if there are model choices available before setting the value
+    if model_choices:
+        return gr.Dropdown.update(choices=model_choices, value=model_choices[0])
+    else:
+        return gr.Dropdown.update(choices=[], value=None)
+        
 with gr.Blocks() as iface:
     gr.Markdown("# Multi-Pipeline Transcription")
     gr.Markdown("Transcribe audio using multiple pipelines and models.")