up

run_demo_ct2.py

@@ -160,23 +160,24 @@ def infer(model, filename, with_timestamps, return_df=False):
             return text
 
 
-def transcribe(microphone, file_upload, with_timestamps, model_name=DEFAULT_MODEL_NAME):
-    warn_output = ""
-    if (microphone is not None) and (file_upload is not None):
-        warn_output = (
-            "WARNING: You've uploaded an audio file and used the microphone. "
-            "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
-        )
+# def transcribe(microphone, file_upload, with_timestamps, model_name=DEFAULT_MODEL_NAME):
+def transcribe(audio_file_path, with_timestamps, model_name=DEFAULT_MODEL_NAME):
+    # warn_output = ""
+    # if (microphone is not None) and (file_upload is not None):
+    #     warn_output = (
+    #         "WARNING: You've uploaded an audio file and used the microphone. "
+    #         "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
+    #     )
 
-    elif (microphone is None) and (file_upload is None):
-        return "ERROR: You have to either use the microphone or upload an audio file"
+    # elif (microphone is None) and (file_upload is None):
+    #     return "ERROR: You have to either use the microphone or upload an audio file"
 
-    file = microphone if microphone is not None else file_upload
+    # audio_file_path = microphone if microphone is not None else file_upload
 
     model = maybe_load_cached_pipeline(model_name)
-    # text = model.transcribe(file, **GEN_KWARGS)["text"]
-    # text = infer(model, file, with_timestamps)
-    text = infer(model, file, with_timestamps, return_df=True)
+    # text = model.transcribe(audio_file_path, **GEN_KWARGS)["text"]
+    # text = infer(model, audio_file_path, with_timestamps)
+    text = infer(model, audio_file_path, with_timestamps, return_df=True)
 
     logger.info(f'Transcription by `{model_name}`:\n{text.to_json(orient="index", force_ascii=False, indent=2)}\n')
 
@@ -237,8 +238,9 @@ with gr.Blocks() as demo:
           """
         )
 
-        microphone_input = gr.Audio(sources="microphone", type="filepath", label="Record")
-        upload_input = gr.Audio(sources="upload", type="filepath", label="Upload File")
+        # microphone_input = gr.Audio(sources="microphone", type="filepath", label="Record")
+        # upload_input = gr.Audio(sources="upload", type="filepath", label="Upload File")
+        audio_file_path = gr.Audio(sources=["microphone", "upload"], type="filepath", label="Record or upload file")
         with_timestamps_input = gr.Checkbox(label="With timestamps?")
 
         microphone_transcribe_btn = gr.Button("Transcribe Audio")
@@ -255,7 +257,7 @@ with gr.Blocks() as demo:
         )
 
         microphone_transcribe_btn.click(
-            transcribe, inputs=[microphone_input, upload_input, with_timestamps_input], outputs=text_output_df2
+            transcribe, inputs=[audio_file_path, with_timestamps_input], outputs=text_output_df2
         )
 
     # with gr.Tab("Transcribe YouTube"):