Update app.py

app.py

@@ -2,11 +2,7 @@ import gradio as gr
 import json
 from faster_whisper import WhisperModel  # Assuming you have installed this library
 
-def split_text_into_lines(data):
-    MaxChars = 30
-    MaxDuration = 2.5
-    MaxGap = 1.5
-
+def split_text_into_lines(data, max_chars, max_duration, max_gap):
     subtitles = []
     line = []
     line_duration = 0
@@ -21,9 +17,9 @@ def split_text_into_lines(data):
 
         temp = " ".join(item["word"] for item in line)
 
-        duration_exceeded = line_duration > MaxDuration
-        chars_exceeded = len(temp) > MaxChars
-        maxgap_exceeded = (word_data['start'] - data[idx - 1]['end']) > MaxGap if idx > 0 else False
+        duration_exceeded = line_duration > max_duration
+        chars_exceeded = len(temp) > max_chars
+        maxgap_exceeded = (word_data['start'] - data[idx - 1]['end']) > max_gap if idx > 0 else False
 
         if duration_exceeded or chars_exceeded or maxgap_exceeded:
             if line:
@@ -48,7 +44,7 @@ def split_text_into_lines(data):
 
     return subtitles
 
-def transcribe_audio(audiofilename):
+def transcribe_audio(audiofilename, max_chars, max_duration, max_gap):
     model_size = "medium"
     model = WhisperModel(model_size)
 
@@ -60,15 +56,18 @@ def transcribe_audio(audiofilename):
         for word in segment.words:
             wordlevel_info.append({'word': word.word, 'start': word.start, 'end': word.end})
 
-    linelevel_subtitles = split_text_into_lines(wordlevel_info)
+    linelevel_subtitles = split_text_into_lines(wordlevel_info, max_chars, max_duration, max_gap)
     return linelevel_subtitles
 
-def audio_transcription(audiofile):
-    transcription = transcribe_audio(audiofile)
+def audio_transcription(audiofile, max_chars, max_duration, max_gap):
+    transcription = transcribe_audio(audiofile, max_chars, max_duration, max_gap)
     return json.dumps(transcription, indent=4)
 
 iface = gr.Interface(audio_transcription, 
-                     gr.Audio(sources="upload", type="filepath"), 
+                     [gr.Audio(sources="upload", type="filepath"),
+                      gr.inputs.Number(label="MaxChars"),
+                      gr.inputs.Number(label="MaxDuration"),
+                      gr.inputs.Number(label="MaxGap")],
                      "text", 
                      description="Upload an audio file and get its transcription in JSON format.")
 iface.launch()