yasserrmd commited on
Commit
bc135da
·
verified ·
1 Parent(s): 901491b

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +9 -4
app.py CHANGED
@@ -3,7 +3,8 @@
3
  import gradio as gr
4
  from extract_text_from_pdf import PDFTextExtractor
5
  from generate_transcript import TranscriptProcessor
6
- from generate_audio import TTSGenerator
 
7
  import pickle
8
  import os
9
  import tempfile
@@ -64,9 +65,12 @@ def process_pdf_to_podcast(pdf_file):
64
 
65
 
66
  # Final Step: Generate Audio after optional adjustments
67
- @spaces.GPU(duration=300)
68
  def generate_audio_from_modified_text(tts_ready_text,session_dir):
69
-
 
 
 
70
  tts_ready_path = os.path.join(session_dir, "podcast_ready_data.pkl")
71
  audio_output_path = os.path.join(session_dir, "final_podcast_audio.mp3")
72
 
@@ -75,7 +79,8 @@ def generate_audio_from_modified_text(tts_ready_text,session_dir):
75
  pickle.dump(tts_ready_text, f)
76
 
77
  # Generate audio from the TTS-ready transcript
78
- tts_gen = TTSGenerator(tts_ready_path,audio_output_path)
 
79
  audio_path = tts_gen.generate_audio()
80
 
81
  return f"Step 4 complete. Audio saved to {audio_path}.", audio_path
 
3
  import gradio as gr
4
  from extract_text_from_pdf import PDFTextExtractor
5
  from generate_transcript import TranscriptProcessor
6
+ #from generate_audio import TTSGenerator
7
+ from generate_audio_edgetts import EdgeTTSGenerator
8
  import pickle
9
  import os
10
  import tempfile
 
65
 
66
 
67
  # Final Step: Generate Audio after optional adjustments
68
+ #@spaces.GPU(duration=300)
69
  def generate_audio_from_modified_text(tts_ready_text,session_dir):
70
+
71
+ if not session_dir:
72
+ session_dir = create_temp_session_directory()
73
+
74
  tts_ready_path = os.path.join(session_dir, "podcast_ready_data.pkl")
75
  audio_output_path = os.path.join(session_dir, "final_podcast_audio.mp3")
76
 
 
79
  pickle.dump(tts_ready_text, f)
80
 
81
  # Generate audio from the TTS-ready transcript
82
+ #tts_gen = TTSGenerator(tts_ready_path,audio_output_path)
83
+ tts_gen = EdgeTTSGenerator(tts_ready_path,audio_output_path)
84
  audio_path = tts_gen.generate_audio()
85
 
86
  return f"Step 4 complete. Audio saved to {audio_path}.", audio_path