WIP

app.py

@@ -20,7 +20,7 @@ def check_ffmpeg():
 # Initialize ffmpeg check
 check_ffmpeg()
 
-def transcribe(inputs):
+def transcribe(inputs, return_timestamps):
     if inputs is None:
         logger.warning("No audio file submitted")
         raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
@@ -39,7 +39,7 @@ def transcribe(inputs):
         
         # Add parameters to request
         params = {
-            "return_timestamps": True
+            "return_timestamps": return_timestamps
         }
         logger.debug(f"Request parameters: {params}")
         
@@ -60,12 +60,12 @@ def transcribe(inputs):
             
         # Format response as JSON
         formatted_result = {
-            "text": result["text"],
-            "chunks": []
+            "text": result["text"]
         }
         
-        if "chunks" in result:
+        if return_timestamps and "chunks" in result:
             logger.info(f"Processing {len(result['chunks'])} chunks")
+            formatted_result["chunks"] = []
             for i, chunk in enumerate(result["chunks"]):
                 logger.debug(f"Processing chunk {i}: {chunk}")
                 try:
@@ -83,14 +83,8 @@ def transcribe(inputs):
                 except Exception as chunk_error:
                     logger.error(f"Error processing chunk {i}: {str(chunk_error)}")
                     continue
-        else:
-            logger.info("No chunks found, using single chunk")
-            formatted_result["chunks"].append({
-                "text": result["text"],
-                "timestamp": [0.0, None]
-            })
+            logger.info(f"Successfully processed transcription with {len(formatted_result['chunks'])} chunks")
         
-        logger.info(f"Successfully processed transcription with {len(formatted_result['chunks'])} chunks")
         return formatted_result
     except Exception as e:
         logger.exception(f"Error during transcription: {str(e)}")
@@ -102,6 +96,7 @@ mf_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
         gr.Audio(sources="microphone", type="filepath"),
+        gr.Checkbox(label="Include timestamps", value=True),
     ],
     outputs=[
         gr.JSON(label="Transcription", open=True),
@@ -123,6 +118,7 @@ file_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
         gr.Audio(sources="upload", type="filepath", label="Audio file"),
+        gr.Checkbox(label="Include timestamps", value=True),
     ],
     outputs=[
         gr.JSON(label="Transcription", open=True),