Create app.py

app.py

@@ -0,0 +1,56 @@
+import torch
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
+import soundfile as sf
+import librosa
+import numpy as np
+from flask import Flask, request, jsonify
+import gradio as gr
+
+app = Flask(__name__)
+
+# Load pre-trained model and tokenizer from Hugging Face
+model_name = "facebook/wav2vec2-large-960h"
+tokenizer = Wav2Vec2Tokenizer.from_pretrained(model_name)
+model = Wav2Vec2ForCTC.from_pretrained(model_name)
+
+def load_audio(file_path):
+    audio, _ = librosa.load(file_path, sr=16000)
+    return audio
+
+def clone_voice(audio):
+    input_values = tokenizer(audio, return_tensors="pt").input_values
+    logits = model(input_values).logits
+    predicted_ids = torch.argmax(logits, dim=-1)
+    transcription = tokenizer.decode(predicted_ids[0])
+
+    # Placeholder for voice conversion logic
+    converted_audio = np.array(audio)  # Replace with actual conversion logic
+
+    output_path = "song_output/output.wav"
+    sf.write(output_path, converted_audio, 16000)
+    return output_path
+
+@app.route('/clone-voice', methods=['POST'])
+def clone_voice_endpoint():
+    if 'file' not in request.files:
+        return jsonify({"error": "No file provided"}), 400
+    
+    file = request.files['file']
+    file_path = "input.wav"
+    file.save(file_path)
+    
+    audio = load_audio(file_path)
+    output_path = clone_voice(audio)
+    
+    return jsonify({"output_path": output_path}), 200
+
+def main_interface(audio):
+    output_path = clone_voice(audio)
+    return output_path
+
+iface = gr.Interface(fn=main_interface, 
+                     inputs=gr.Audio(source="upload", type="numpy"), 
+                     outputs=gr.Audio(type="file"))
+
+if __name__ == "__main__":
+    iface.launch(server_name="0.0.0.0", server_port=5000)