Spaces:
Paused
Paused
| import logging | |
| import wave | |
| import numpy as np | |
| from io import BytesIO | |
| from flask import Flask, request, send_file, jsonify | |
| from flask_cors import CORS | |
| from huggingface_hub import hf_hub_download | |
| from piper import PiperVoice | |
| app = Flask(__name__) | |
| CORS(app) | |
| # Setup logging | |
| logging.basicConfig(level=logging.DEBUG) | |
| logger = logging.getLogger(__name__) | |
| def synthesize_speech(text, sentence_silence, length_scale): | |
| logger.debug("Downloading model and config files...") | |
| model_path = hf_hub_download(repo_id="csukuangfj/vits-piper-en_US-lessac-medium", filename="en_US-lessac-medium.onnx") | |
| config_path = hf_hub_download(repo_id="csukuangfj/vits-piper-en_US-lessac-medium", filename="en_US-lessac-medium.onnx.json") | |
| logger.debug("Loading PiperVoice model...") | |
| voice = PiperVoice.load(model_path, config_path) | |
| buffer = BytesIO() | |
| logger.debug("Synthesizing speech...") | |
| with wave.open(buffer, 'wb') as wav_file: | |
| wav_file.setframerate(voice.config.sample_rate) | |
| wav_file.setsampwidth(2) | |
| wav_file.setnchannels(1) | |
| voice.synthesize(text, wav_file, sentence_silence=sentence_silence, length_scale=length_scale) | |
| buffer.seek(0) | |
| logger.debug("Speech synthesis complete.") | |
| return buffer | |
| def index(): | |
| return ''' | |
| <!DOCTYPE html> | |
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>TTS Server</title> | |
| <style> | |
| body { font-family: Arial, sans-serif; margin: 40px; } | |
| h1 { color: #333; } | |
| p { font-size: 1.2em; } | |
| code { background: #f4f4f4; padding: 2px 4px; border-radius: 4px; } | |
| </style> | |
| </head> | |
| <body> | |
| <h1>TTS Server is Running</h1> | |
| <p>Use the <code>/tts</code> endpoint to synthesize speech.</p> | |
| <p>Send a POST request with JSON data containing the <code>text</code>, <code>sentence_silence</code>, and <code>length_scale</code> parameters.</p> | |
| </body> | |
| </html> | |
| ''' | |
| def tts(): | |
| data = request.json | |
| if not data: | |
| logger.error("No data received in request.") | |
| return jsonify({"error": "No data provided"}), 400 | |
| text = data.get('text', '') | |
| sentence_silence = float(data.get('sentence_silence', 0.1)) | |
| length_scale = float(data.get('length_scale', 1.0)) | |
| if not text: | |
| logger.error("No text provided in request.") | |
| return jsonify({"error": "Text parameter is required"}), 400 | |
| logger.info(f"Received request: text={text}, sentence_silence={sentence_silence}, length_scale={length_scale}") | |
| try: | |
| audio_buffer = synthesize_speech(text, sentence_silence, length_scale) | |
| except Exception as e: | |
| logger.exception("Error during speech synthesis.") | |
| return jsonify({"error": str(e)}), 500 | |
| return send_file(audio_buffer, mimetype="audio/wav", as_attachment=True, download_name="output.wav") | |
| if __name__ == '__main__': | |
| app.run(debug=True, host='0.0.0.0', port=7860) | |