Update app.py

app.py

@@ -6,21 +6,23 @@ import zipfile
 from TTS.api import TTS
 from pydub import AudioSegment
 
-os.environ["COQUI_TOS_AGREED"] = "1"
-
-MODEL_PATH = "tts_models/multilingual/multi-dataset/xtts_v2"
-LANGUAGES = ["en", "es", "fr", "de", "it", "pt", "pl", "tr", "ru", "nl", "cs", "ar", "zh-cn", "ja", "hu", "ko", "hi"]
+# Constants
 AUDIO_FORMATS = [".wav", ".mp3", ".flac", ".mp4"]
+LANGUAGES = ["en", "es", "fr", "de", "it", "pt", "pl", "tr", "ru", "nl", "cs", "ar", "zh-cn", "ja", "hu", "ko", "hi"]
 
+# Device setup
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 print(f"Using device: {device}")
 
-tts = TTS(MODEL_PATH).to(device)
+# TTS model setup
+tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)
 
-def download_audio_file(url):
+def download_audio_file(url: str) -> str:
     try:
         response = requests.get(url)
         file_extension = os.path.splitext(url)[-1].lower()
+        if file_extension not in AUDIO_FORMATS:
+            raise ValueError(f"Unsupported file extension: {file_extension}")
         file_name = f"temp{file_extension}"
         with open(file_name, "wb") as f:
             f.write(response.content)
@@ -29,7 +31,7 @@ def download_audio_file(url):
         print(f"Error downloading audio file: {e}")
         return None
 
-def extract_zip_file(zip_file):
+def extract_zip_file(zip_file: str) -> bool:
     try:
         with zipfile.ZipFile(zip_file, 'r') as zip_ref:
             zip_ref.extractall()
@@ -38,52 +40,48 @@ def extract_zip_file(zip_file):
         print(f"Error extracting zip file: {e}")
         return False
 
-def convert_to_wav(input_audio_file):
-    file_extension = os.path.splitext(input_audio_file)[-1].lower()
-    if file_extension!= ".wav":
-        audio = AudioSegment.from_file(input_audio_file)
-        audio.export("temp.wav", format="wav")
-        os.remove(input_audio_file)
-        return "temp.wav"
-    return input_audio_file
+def audio_processing:
+    def convert_to_wav(input_audio_file: str) -> str:
+        file_extension = os.path.splitext(input_audio_file)[-1].lower()
+        if file_extension!= ".wav":
+            audio = AudioSegment.from_file(input_audio_file)
+            audio.export("temp.wav", format="wav")
+            os.remove(input_audio_file)
+            return "temp.wav"
+        return input_audio_file
 
-def synthesize_text(text, input_audio_file, language):
-    input_audio_file = convert_to_wav(input_audio_file)
-    tts.tts_to_file(text=text, speaker_wav=input_audio_file, language=language, file_path="./output.wav")
-    return "./output.wav"
+    def synthesize_text(text: str, input_audio_file: str, language: str) -> str:
+        input_audio_file = convert_to_wav(input_audio_file)
+        tts.tts_to_file(text=text, speaker_wav=input_audio_file, language=language, file_path="./output.wav")
+        return "./output.wav"
 
-def clone(text, input_file, language, url=None):
-    if url is not None:
-        input_file = download_audio_file(url)
-    if input_file is None:
-        return None
-    if input_file.name.endswith(".zip"):
-        if extract_zip_file(input_file):
-            input_audio_file = [f for f in os.listdir('.') if os.path.isfile(f) and f.endswith(tuple(AUDIO_FORMATS))]
-            if len(input_audio_file) == 1:
-                input_audio_file = input_audio_file[0]
-            else:
-                return "Error: Please select a single audio file from the extracted files."
+def voice_clone(text: str, input_file: gr.File, language: str, url: gr.Audio) -> str:
+    if input_file is not None:
+        if input_file.name.endswith(".zip"):
+            if extract_zip_file(input_file):
+                input_audio_file = [f for f in os.listdir('.') if os.path.isfile(f) and f.endswith(tuple(AUDIO_FORMATS))]
+                if len(input_audio_file) == 1:
+                    input_audio_file = input_audio_file[0]
+                else:
+                    return "Error: Please select a single audio file from the extracted files."
+        else:
+            input_audio_file = input_file.name
+    elif url:
+        input_audio_file = url.name
     else:
-        input_audio_file = input_file.name
+        return"Error: Please select a file or enter a URL"
+    if language not in LANGUAGES:
+        return "Error: Invalid language selected"
     output_file_path = synthesize_text(text, input_audio_file, language)
     return output_file_path
 
 iface = gr.Interface(
-    fn=clone,
-    inputs=[
-        gr.inputs.Textbox(label="Text"),
-        gr.inputs.File(label="Input File", file_types=[".zip", *AUDIO_FORMATS]),
-        gr.inputs.Dropdown(choices=LANGUAGES, label="Language"),
-        gr.inputs.Textbox(label="URL", lines=1)
-    ],
-    outputs=gr.outputs.Audio(type='filepath'),
+    fn=voice_clone,
+    inputs=["text", gr.File(label="Input File", file_types=[".zip", *AUDIO_FORMATS]), gr.Dropdown(choices=LANGUAGES, label="Language"), gr.Audio(label="URL")],
+    outputs=gr.Audio(type='filepath'),
     title='Voice Clone',
-    description="""
-    by [Angetyde](https://youtube.com/@Angetyde) 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")
+    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="green"),
 )
 
-iface.launch(share=True)
+iface.launch()