Update app.py

app.py

@@ -5,7 +5,6 @@ import zipfile
 import requests
 from TTS.api import TTS
 from pydub import AudioSegment
-import tempfile
 
 # Set environment variable
 os.environ["COQUI_TOS_AGREED"] = "1"
@@ -22,51 +21,51 @@ print(f"Using device: {device}")
 # Load TTS model
 tts = TTS(MODEL_PATH).to(device)
 
-def download_audio_from_url(url: str) -> str:
- try:
- response = requests.get(url)
- response.raise_for_status()
- with tempfile.NamedTemporaryFile(suffix=".wav") as tmp:
- tmp.write(response.content)
- return tmp.name
- except requests.RequestException as e:
- print(f"Error downloading audio: {e}")
- return None
-
-def convert_audio_to_wav(audio_file: str) -> str:
+def convert_to_wav(audio_file):
+ # Extract file extension
  file_extension = os.path.splitext(audio_file)[-1].lower()
- if file_extension!= ".wav":
+
+ # Convert audio file to .wav format
+ if file_extension != ".wav":
  audio = AudioSegment.from_file(audio_file)
- with tempfile.NamedTemporaryFile(suffix=".wav") as tmp:
- audio.export(tmp.name, format="wav")
- return tmp.name
+ audio.export("temp.wav", format="wav")
+ os.remove(audio_file)
+ audio_file = "temp.wav"
+
  return audio_file
 
-def generate_audio(text: str, audio_file: str, language: str) -> str:
- try:
- tts.tts_to_file(text=text, speaker_wav=audio_file, language=language, file_path="./output.wav")
- return "./output.wav"
- except Exception as e:
- print(f"Error generating audio: {e}")
- return None
+def clone(text, url, language):
+ # Download zip file
+ response = requests.get(url)
+ with open("temp.zip", "wb") as f:
+ f.write(response.content)
+
+ # Extract audio file from zip archive
+ with zipfile.ZipFile("temp.zip", "r") as zip_ref:
+ for file in zip_ref.namelist():
+ if os.path.splitext(file)[-1].lower() in AUDIO_FORMATS:
+ zip_ref.extract(file, ".")
+ audio_file = file
+ break
+
+ # Convert audio file to .wav format
+ audio_file = convert_to_wav(audio_file)
+
+ # Generate audio using TTS model
+ tts.tts_to_file(text=text, speaker_wav=audio_file, language=language, file_path="./output.wav")
 
-def clone(text: str, audio_or_url: str, language: str) -> str:
- if isinstance(audio_or_url, str):
- audio_file = download_audio_from_url(audio_or_url)
- else:
- with tempfile.NamedTemporaryFile(suffix=".wav") as tmp:
- tmp.write(audio_or_url.read())
- audio_file = tmp.name
+ # Clean up
+ os.remove(audio_file)
+ os.remove("temp.zip")
 
- audio_file = convert_audio_to_wav(audio_file)
- return generate_audio(text, audio_file, language)
+ return "./output.wav"
 
 # Create Gradio interface
 iface = gr.Interface(
  fn=clone,
- inputs=["text", gr.Audio(label="Upload audio file or enter URL"), gr.Dropdown(choices=LANGUAGES, label="Language")],
- output_type="filepath",
- title="Voice Clone",
+ inputs=["text", gr.components.Text(label="URL"), gr.Dropdown(choices=LANGUAGES, label="Language")],
+ outputs=gr.Audio(type='filepath'),
+ title='Voice Clone',
  description=""" by [Angetyde](https://youtube.com/@Angetyde?si=7nusP31nTumIkPTF) and [Tony Assi](https://www.tonyassi.com/ ) use this colab with caution <3. """,
  theme=gr.themes.Base(primary_hue="teal", secondary_hue="teal", neutral_hue="slate")
 )