Upload 3 files

README.md

@@ -1,12 +1,72 @@
 ---
-title: Kani Tts Web App
-emoji: 🦀
-colorFrom: purple
-colorTo: green
+title: KaniTTS Web App
+emoji: 🎤
+colorFrom: blue
+colorTo: purple
 sdk: gradio
-sdk_version: 6.0.2
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
+license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# KaniTTS Web App
+
+A text-to-speech web application using KaniTTS model with Gradio interface.
+
+## Prerequisites
+
+- Python 3.10 or higher
+- PyTorch 2.8.0 or higher (requires Linux or Apple Silicon Mac)
+
+> **Note**: This app requires PyTorch 2.8+. macOS x86_64 (Intel) only supports up to PyTorch 2.2.2. For Intel Macs, please use a cloud platform.
+
+## Cloud Deployment
+
+### Google Colab
+1. Upload the files to Google Drive or clone from GitHub
+2. Run: `!pip install -r requirements.txt`
+3. Run: `!python app.py`
+
+### Hugging Face Spaces
+1. Create a new Space with Gradio SDK
+2. Upload `app.py` and `requirements.txt`
+3. The app will automatically deploy
+
+### Replit
+1. Create a new Python Repl
+2. Upload the files
+3. Install dependencies: `pip install -r requirements.txt`
+4. Run: `python app.py`
+
+## Local Installation (Linux/Apple Silicon)
+
+```bash
+# Clone the repository
+git clone <your-repo-url>
+cd tts-app
+
+# Create virtual environment
+python3 -m venv venv
+source venv/bin/activate
+
+# Install dependencies
+pip install -r requirements.txt
+
+# Run the application
+python app.py
+```
+
+## Usage
+
+1. Open your browser and go to: `http://127.0.0.1:7860`
+2. Enter the text you want to convert to speech
+3. Click the "Generate" button
+4. Listen to the generated audio
+
+## Model Information
+
+- **Model**: `nineninesix/kani-tts-400m-en`
+- **Sample Rate**: 22050Hz (model default)
+- **Language**: English
+

app.py

@@ -0,0 +1,57 @@
+import gradio as gr
+from kani_tts import KaniTTS
+import soundfile as sf
+import numpy as np
+import tempfile
+import os
+
+# Initialize the KaniTTS model
+print("Loading KaniTTS model...")
+model = KaniTTS('nineninesix/kani-tts-400m-en')
+print("Model loaded successfully!")
+
+def generate_speech(text):
+    """Generate speech from text using KaniTTS model."""
+    if not text or not text.strip():
+        return None
+    
+    try:
+        # Generate audio from text
+        audio, _ = model(text)
+        
+        # Get sample rate from model (typically 22050Hz or 24000Hz)
+        sample_rate = model.sample_rate
+        
+        # Save to temporary file
+        temp_file = tempfile.NamedTemporaryFile(suffix=".wav", delete=False)
+        sf.write(temp_file.name, audio, sample_rate)
+        
+        return temp_file.name
+    except Exception as e:
+        print(f"Error generating speech: {e}")
+        return None
+
+# Create Gradio interface
+with gr.Blocks(title="KaniTTS Web App") as demo:
+    gr.Markdown("# KaniTTS Web App")
+    gr.Markdown("Enter text below and click Generate to convert it to speech.")
+    
+    with gr.Row():
+        text_input = gr.Textbox(
+            label="Text Input",
+            placeholder="Enter text to convert to speech...",
+            lines=3
+        )
+    
+    generate_btn = gr.Button("Generate", variant="primary")
+    
+    audio_output = gr.Audio(label="Generated Audio", type="filepath")
+    
+    generate_btn.click(
+        fn=generate_speech,
+        inputs=text_input,
+        outputs=audio_output
+    )
+
+if __name__ == "__main__":
+    demo.launch(server_port=7860)

requirements.txt

@@ -0,0 +1,6 @@
+kani-tts>=0.0.4
+gradio>=4.0.0
+torch>=2.8.0
+soundfile>=0.12.0
+numpy>=1.24.0
+transformers>=4.57.0