Update app.py

app.py

@@ -1,7 +1,5 @@
 import gradio as gr
-import torch
 import os
-import zipfile
 import requests
 from TTS.api import TTS
 from pydub import AudioSegment
@@ -17,28 +15,6 @@ print(f"Using device: {device}")
 
 tts = TTS(MODEL_PATH).to(device)
 
-def download_zip_file(url):
- try:
- response = requests.get(url)
- with open("temp.zip", "wb") as f:
- f.write(response.content)
- return "temp.zip"
- except requests.exceptions.RequestException as e:
- print(f"Error downloading ZIP file: {e}")
- return None
-
-def extract_audio_file(zip_file_path):
- try:
- with zipfile.ZipFile(zip_file_path, "r") as zip_ref:
- for file in zip_ref.namelist():
- if os.path.splitext(file)[-1].lower() in AUDIO_FORMATS:
- zip_ref.extract(file, ".")
- return file
- return None
- except zipfile.BadZipFile as e:
- print(f"Error extracting ZIP file: {e}")
- return None
-
 def convert_to_wav(input_audio_file):
  file_extension = os.path.splitext(input_audio_file)[-1].lower()
  if file_extension!= ".wav":
@@ -53,20 +29,16 @@ def synthesize_text(text, input_audio_file, language):
  tts.tts_to_file(text=text, speaker_wav=input_audio_file, language=language, file_path="./output.wav")
  return "./output.wav"
 
-def clone(text, url, language):
- zip_file_path = download_zip_file(url)
- if zip_file_path:
- input_audio_file = extract_audio_file(zip_file_path)
- if input_audio_file:
- output_file_path = synthesize_text(text, input_audio_file, language)
- os.remove(input_audio_file)
- os.remove(zip_file_path)
- return output_file_path
- return None
+def clone(text, input_file, language):
+ if input_file is None:
+ return None
+ input_audio_file = input_file.name
+ output_file_path = synthesize_text(text, input_audio_file, language)
+ return output_file_path
 
 iface = gr.Interface(
  fn=clone,
- inputs=["text", gr.Textbox(label="URL"), gr.Dropdown(choices=LANGUAGES, label="Language")],
+ inputs=["text", gr.File(label="Input File", file_types=AUDIO_FORMATS), 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. """,