Update app.py

app.py

@@ -1,6 +1,7 @@
 import gradio as gr
 from TTS.api import TTS
 import tempfile
+import os
 
 # Load multilingual TTS model (supports English and Arabic, with voice options)
 model_name = "tts_models/multilingual/multi-dataset/your_tts"
@@ -9,26 +10,86 @@ tts = TTS(model_name)
 # Get available speakers from model
 available_speakers = tts.speakers
 
-def text_to_speech(text, language, speaker_name):
+def text_to_speech(text, language, speaker_name, speed, pitch):
     try:
+        # Validate inputs
+        if not text.strip():
+            raise ValueError("Please enter some text")
+        
+        # Create parameters dictionary with optional parameters
+        params = {
+            "text": text,
+            "speaker": speaker_name,
+            "language": language,
+            "file_path": None  # We'll handle the file ourselves
+        }
+        
+        # Add optional parameters if they're not default values
+        if speed != 1.0:
+            params["speed"] = speed
+        if pitch != 1.0:
+            params["pitch"] = pitch
+        
         # Save to temporary WAV file
         with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as f:
-            tts.tts_to_file(text=text, speaker=speaker_name, language=language, file_path=f.name)
+            params["file_path"] = f.name
+            tts.tts_to_file(**params)
             return f.name
+        
     except Exception as e:
-        return f"Error: {str(e)}"
+        raise gr.Error(f"Error generating speech: {str(e)}")
 
-interface = gr.Interface(
-    fn=text_to_speech,
-    inputs=[
-        gr.Textbox(label="Enter text (English or Arabic)"),
-        gr.Dropdown(choices=["en", "ar"], label="Language"),
-        gr.Dropdown(choices=available_speakers, label="Voice (Male/Female)"),
-    ],
-    outputs=gr.Audio(type="filepath", label="Generated Audio"),
-    title="English & Arabic Text-to-Speech App",
-    description="Type your text, choose the language and voice, and listen!"
-)
+def create_download_link(audio_file):
+    if audio_file is None or not os.path.exists(audio_file):
+        return None
+    return gr.DownloadButton(label="Download Audio", value=audio_file)
+
+with gr.Blocks(title="Enhanced TTS App") as app:
+    gr.Markdown("# Enhanced English & Arabic Text-to-Speech")
+    gr.Markdown("Type your text, adjust settings, and generate speech!")
+    
+    with gr.Row():
+        with gr.Column():
+            text_input = gr.Textbox(label="Enter text (English or Arabic)", lines=5)
+            language = gr.Dropdown(choices=["en", "ar"], label="Language", value="en")
+            speaker = gr.Dropdown(choices=available_speakers, label="Voice (Male/Female)")
+            
+            with gr.Accordion("Advanced Settings", open=False):
+                speed = gr.Slider(minimum=0.5, maximum=2.0, value=1.0, step=0.1, 
+                                 label="Speed (1.0 = normal)")
+                pitch = gr.Slider(minimum=0.5, maximum=2.0, value=1.0, step=0.1, 
+                                label="Pitch (1.0 = normal)")
+            
+            generate_btn = gr.Button("Generate Speech", variant="primary")
+        
+        with gr.Column():
+            audio_output = gr.Audio(label="Generated Audio", type="filepath")
+            download_section = gr.Group(visible=False)
+    
+    # Set up interactivity
+    generate_btn.click(
+        fn=text_to_speech,
+        inputs=[text_input, language, speaker, speed, pitch],
+        outputs=audio_output
+    ).then(
+        fn=lambda: gr.Group(visible=True),
+        outputs=download_section
+    ).then(
+        fn=create_download_link,
+        inputs=audio_output,
+        outputs=download_section
+    )
+    
+    gr.Examples(
+        examples=[
+            ["Hello, welcome to our text-to-speech application!", "en", available_speakers[0], 1.0, 1.0],
+            ["مرحبا بكم في تطبيق تحويل النص إلى كلام", "ar", available_speakers[-1], 1.0, 1.0]
+        ],
+        inputs=[text_input, language, speaker, speed, pitch],
+        outputs=audio_output,
+        fn=text_to_speech,
+        cache_examples=True
+    )
 
 if __name__ == "__main__":
-    interface.launch()
+    app.launch()