WIP

app.py

@@ -276,8 +276,8 @@ def transcribe_youtube(url, return_timestamps, generate_subs, chunk_length_s=30,
             "text": result["transcription"]["text"],
             "segments": [
                 {
-                    "start": segment["start"],
-                    "end": segment["end"],
+                    "start": segment["start_time"],
+                    "end": segment["end_time"],
                     "text": segment["text"]
                 }
                 for segment in result["transcription"]["segments"]
@@ -347,8 +347,8 @@ def transcribe(inputs, return_timestamps, generate_subs, chunk_length_s=30, batc
                 "text": result["transcription"]["text"],
                 "segments": [
                     {
-                        "start": segment["start"],
-                        "end": segment["end"],
+                        "start": segment["start_time"],
+                        "end": segment["end_time"],
                         "text": segment["text"]
                     }
                     for segment in result["transcription"]["segments"]
@@ -384,7 +384,7 @@ youtube_transcribe = gr.Interface(
         gr.Textbox(label="YouTube URL", placeholder="https://www.youtube.com/watch?v=..."),
         gr.Checkbox(label="Include timestamps", value=True),
         gr.Checkbox(label="Generate subtitles", value=True),
-        gr.Slider(minimum=10, maximum=60, value=30, step=5, label="Chunk Length (seconds)"),
+        gr.Slider(minimum=5, maximum=30, value=15, step=5, label="Chunk Length (seconds)"),
         gr.Slider(minimum=32, maximum=256, value=128, step=32, label="Batch Size")
     ],
     outputs=[
@@ -408,7 +408,7 @@ mf_transcribe = gr.Interface(
         gr.Audio(sources="microphone", type="filepath"),
         gr.Checkbox(label="Include timestamps", value=True),
         gr.Checkbox(label="Generate subtitles", value=True),
-        gr.Slider(minimum=10, maximum=30, value=30, step=5, label="Chunk Length (seconds)"),
+        gr.Slider(minimum=5, maximum=30, value=30, step=5, label="Chunk Length (seconds)"),
         gr.Slider(minimum=32, maximum=256, value=128, step=32, label="Batch Size")
     ],
     outputs=[
@@ -429,7 +429,7 @@ file_transcribe = gr.Interface(
         gr.Audio(sources="upload", type="filepath", label="Audio file"),
         gr.Checkbox(label="Include timestamps", value=True),
         gr.Checkbox(label="Generate subtitles", value=True),
-        gr.Slider(minimum=10, maximum=30, value=15, step=5, label="Chunk Length (seconds)"),
+        gr.Slider(minimum=5, maximum=30, value=15, step=5, label="Chunk Length (seconds)"),
         gr.Slider(minimum=32, maximum=256, value=8, step=32, label="Batch Size")
     ],
     outputs=[