Update app.py

app.py

@@ -10,7 +10,12 @@ from gtts.lang import _main_langs
 AUDIO_DIR = 'audio_files'
 MAX_FILE_AGE = 24 * 60 * 60  # maximum age of audio files in seconds (24 hours)
 
-async def text_to_speech(text, lang, tld, method):
+# Lấy danh sách giọng nói cho Edge TTS
+async def get_edge_voices():
+    voices = await edge_tts.get_voices()
+    return [voice['name'] for voice in voices]
+
+async def text_to_speech(text, lang, tld, method, voice):
     lang_codes = {lang_name: lang_code for lang_code, lang_name in _main_langs().items()}
     lang_code = lang_codes[lang]
 
@@ -23,7 +28,6 @@ async def text_to_speech(text, lang, tld, method):
     elif method == "Edge TTS":
         # Convert text to speech using Edge TTS
         output = io.BytesIO()
-        voice = f"{lang_code}-{tld}".replace('_', '-')
         # Create an Edge TTS object and get audio
         communicate = edge_tts.Communicate(text, voice)
         await communicate.save(output)
@@ -49,37 +53,43 @@ def delete_old_audio_files():
             os.remove(file_path)
 
 # Hàm chuyển đổi file .txt thành giọng nói
-async def txt_to_speech(file, lang, tld, method):
+async def txt_to_speech(file, lang, tld, method, voice):
     with open(file.name, 'r') as f:
         text = f.read()
-    return await text_to_speech(text, lang, tld, method)
-
-tlds = [
-    "com", "ad", "ae", "com.af", "com.ag", "com.ai", "com.ar", "as", "at", "com.au", "az", 
-    # (remaining TLDs omitted for brevity)
-    "com.vn"
-]
+    return await text_to_speech(text, lang, tld, method, voice)
 
 # Tạo giao diện Gradio với tab
 with gr.Blocks() as iface:
+    voices = asyncio.run(get_edge_voices())
+    
     with gr.Tab("Text to Speech"):
         gr.Markdown("### Convert text to speech")
         text_input = gr.Textbox(lines=10, label="Enter your text here:")
         lang_input = gr.Dropdown(choices=list(_main_langs().values()), label="Select language:")
-        tld_input = gr.Dropdown(choices=[tld for tld in tlds], label="Select TLD:", value="com")
+        tld_input = gr.Dropdown(choices=tlds, label="Select TLD:", value="com")
         method_input = gr.Radio(choices=["gTTS", "Edge TTS"], label="Select method:", value="gTTS")
+        
+        # Dropdown cho giọng nói
+        voice_input = gr.Dropdown(choices=voices, label="Select voice:", value=voices[0])
+        
         audio_output, file_output = gr.Audio(label="Audio"), gr.File(label="Audio File")
-        gr.Button("Convert").click(fn=lambda text, lang, tld, method: asyncio.run(text_to_speech(text, lang, tld, method)), 
-                                    inputs=[text_input, lang_input, tld_input, method_input], outputs=[audio_output, file_output])
+        gr.Button("Convert").click(fn=lambda text, lang, tld, method, voice: asyncio.run(text_to_speech(text, lang, tld, method, voice)),
+                                    inputs=[text_input, lang_input, tld_input, method_input, voice_input], 
+                                    outputs=[audio_output, file_output])
     
     with gr.Tab("TXT to Speech"):
         gr.Markdown("### Convert .txt file to speech")
         file_input = gr.File(label="Upload your .txt file")
         lang_input_file = gr.Dropdown(choices=list(_main_langs().values()), label="Select language:")
-        tld_input_file = gr.Dropdown(choices=[tld for tld in tlds], label="Select TLD:", value="com")
+        tld_input_file = gr.Dropdown(choices=tlds, label="Select TLD:", value="com")
         method_input_file = gr.Radio(choices=["gTTS", "Edge TTS"], label="Select method:", value="gTTS")
+        
+        # Dropdown cho giọng nói
+        voice_input_file = gr.Dropdown(choices=voices, label="Select voice:", value=voices[0])
+        
         audio_output_file, file_output_file = gr.Audio(label="Audio"), gr.File(label="Audio File")
-        gr.Button("Convert").click(fn=lambda file, lang, tld, method: asyncio.run(txt_to_speech(file, lang, tld, method)), 
-                                    inputs=[file_input, lang_input_file, tld_input_file, method_input_file], outputs=[audio_output_file, file_output_file])
+        gr.Button("Convert").click(fn=lambda file, lang, tld, method, voice: asyncio.run(txt_to_speech(file, lang, tld, method, voice)),
+                                    inputs=[file_input, lang_input_file, tld_input_file, method_input_file, voice_input_file], 
+                                    outputs=[audio_output_file, file_output_file])
 
 iface.launch(enable_queue=True)