simonkoson
d6edc69c65
- Flask backend with MiniMax and CosyVoice TTS engines - Frontend UI with emotion tags, speed/pitch/volume controls - Nginx reverse proxy configuration - Dual engine support (MiniMax speech-2.8-hd, CosyVoice v3.5-flash)
58 lines
1.7 KiB
Python
58 lines
1.7 KiB
Python
from flask import Blueprint, request, jsonify
|
|
import base64
|
|
import os
|
|
import requests
|
|
from app.services.minimax_service import synthesize_minimax
|
|
from app.services.cosyvoice_service import synthesize_cosyvoice
|
|
|
|
bp = Blueprint('tts', __name__, url_prefix='/api')
|
|
|
|
MINIMAX_API_KEY = os.getenv('MINIMAX_API_KEY', '')
|
|
MINIMAX_API_URL = 'https://api.minimax.chat/v1/t2a_v2'
|
|
|
|
COSYVOICE_API_URL = os.getenv('COSYVOICE_API_URL', 'http://127.0.0.1:5001/v1/audio/synthesis')
|
|
|
|
@bp.route('/synthesize', methods=['POST'])
|
|
def synthesize():
|
|
data = request.get_json()
|
|
|
|
if not data or 'text' not in data:
|
|
return jsonify({'error': '缺少text参数'}), 400
|
|
|
|
text = data.get('text')
|
|
engine = data.get('engine', 'minimax')
|
|
speed = data.get('speed', 1.0)
|
|
pitch = data.get('pitch', 0)
|
|
volume = data.get('volume', 1.0)
|
|
instruction = data.get('instruction') # 用于CosyVoice的情绪指令
|
|
|
|
if not text.strip():
|
|
return jsonify({'error': '文本不能为空'}), 400
|
|
|
|
if len(text) > 500:
|
|
return jsonify({'error': '文本不能超过500字'}), 400
|
|
|
|
try:
|
|
if engine == 'cosyvoice':
|
|
audio_data = synthesize_cosyvoice(
|
|
text=text,
|
|
speed=speed,
|
|
pitch=pitch,
|
|
volume=volume,
|
|
instruction=instruction
|
|
)
|
|
else:
|
|
audio_data = synthesize_minimax(
|
|
text=text,
|
|
speed=speed,
|
|
pitch=pitch,
|
|
volume=volume
|
|
)
|
|
|
|
return jsonify({
|
|
'success': True,
|
|
'audio': base64.b64encode(audio_data).decode('utf-8')
|
|
})
|
|
except Exception as e:
|
|
return jsonify({'error': str(e)}), 500
|