Elieon commited on
Commit
88b8d7c
·
verified ·
1 Parent(s): fad23ac

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +8 -11
app.py CHANGED
@@ -10,15 +10,15 @@ async def get_voices():
10
  return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
11
 
12
  # Text-to-speech function
13
- async def text_to_speech(text, voice, rate, pitch):
14
  if not text.strip():
15
  return None, gr.Warning("Please enter text to convert.")
16
  if not voice:
17
  return None, gr.Warning("Please select a voice.")
18
 
19
  voice_short_name = voice.split(" - ")[0]
20
- rate_str = f"{rate:+d}%"
21
- pitch_str = f"{pitch:+d}Hz"
22
  communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
23
  with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
24
  tmp_path = tmp_file.name
@@ -26,13 +26,12 @@ async def text_to_speech(text, voice, rate, pitch):
26
  return tmp_path, None
27
 
28
  # Gradio interface function
29
- def tts_interface(text, voice, rate, pitch):
30
- audio, warning = asyncio.run(text_to_speech(text, voice, rate, pitch))
 
31
  return audio, warning
32
 
33
  # Create Gradio application
34
- import gradio as gr
35
-
36
  async def create_demo():
37
  voices = await get_voices()
38
 
@@ -40,9 +39,7 @@ async def create_demo():
40
  fn=tts_interface,
41
  inputs=[
42
  gr.Textbox(label="Input Text", lines=5),
43
- gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value=""),
44
- gr.Slider(minimum=-50, maximum=50, value=0, label="Speech Rate Adjustment (%)", step=1),
45
- gr.Slider(minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1)
46
  ],
47
  outputs=[
48
  gr.Audio(label="Generated Audio", type="filepath"),
@@ -57,4 +54,4 @@ async def create_demo():
57
  # Run the application
58
  if __name__ == "__main__":
59
  demo = asyncio.run(create_demo())
60
- demo.launch()
 
10
  return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
11
 
12
  # Text-to-speech function
13
+ async def text_to_speech(text, voice, rate=0, pitch=0):
14
  if not text.strip():
15
  return None, gr.Warning("Please enter text to convert.")
16
  if not voice:
17
  return None, gr.Warning("Please select a voice.")
18
 
19
  voice_short_name = voice.split(" - ")[0]
20
+ rate_str = f"{rate:+d}%" # Default to 0% for rate
21
+ pitch_str = f"{pitch:+d}Hz" # Default to 0Hz for pitch
22
  communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
23
  with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
24
  tmp_path = tmp_file.name
 
26
  return tmp_path, None
27
 
28
  # Gradio interface function
29
+ def tts_interface(text, voice):
30
+ # Pass default values for rate and pitch
31
+ audio, warning = asyncio.run(text_to_speech(text, voice, rate=0, pitch=0))
32
  return audio, warning
33
 
34
  # Create Gradio application
 
 
35
  async def create_demo():
36
  voices = await get_voices()
37
 
 
39
  fn=tts_interface,
40
  inputs=[
41
  gr.Textbox(label="Input Text", lines=5),
42
+ gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value="")
 
 
43
  ],
44
  outputs=[
45
  gr.Audio(label="Generated Audio", type="filepath"),
 
54
  # Run the application
55
  if __name__ == "__main__":
56
  demo = asyncio.run(create_demo())
57
+ demo.launch()