nanocode/audio-recorder/backend/app_v2.py

#!/usr/bin/env python3
"""
录音后端服务 V2 - 集成语音转文字
"""
import warnings
warnings.filterwarnings('ignore')

from flask import Flask, request, jsonify, send_from_directory
from flask_cors import CORS
from werkzeug.utils import secure_filename
import os
import ssl
from datetime import datetime

# 导入转写模块 - 使用 Vosk
from transcribe_api import transcribe_audio

MODEL_PATH = '/home/ubuntu/nanocode/audio-recorder/models/cn-model'

app = Flask(__name__)
CORS(app)

# 配置
UPLOAD_FOLDER = '/home/ubuntu/nanocode/audio-recorder/uploads'
ALLOWED_EXTENSIONS = {'webm', 'mp4', 'wav', 'mp3', 'ogg'}
MAX_CONTENT_LENGTH = 100 * 1024 * 1024

app.config['MAX_CONTENT_LENGTH'] = MAX_CONTENT_LENGTH
os.makedirs(UPLOAD_FOLDER, exist_ok=True)


def allowed_file(filename):
    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS


@app.route('/api/upload', methods=['POST'])
def upload_audio():
    """接收音频文件上传"""
    if 'audio' not in request.files:
        return jsonify({'error': '没有文件'}), 400
    
    file = request.files['audio']
    if file.filename == '':
        return jsonify({'error': '文件名为空'}), 400
    
    if file and allowed_file(file.filename):
        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
        filename = f"recording_{timestamp}.webm"
        filepath = os.path.join(UPLOAD_FOLDER, filename)
        
        file.save(filepath)
        file_size = os.path.getsize(filepath)
        
        return jsonify({
            'success': True,
            'filename': filename,
            'filepath': filepath,
            'size': file_size,
            'size_human': f"{file_size / 1024:.1f} KB",
            'timestamp': timestamp
        }), 200
    
    return jsonify({'error': '不支持的文件类型'}), 400


@app.route('/api/transcribe', methods=['POST'])
def transcribe():
    """语音转文字"""
    if 'audio' not in request.files:
        return jsonify({'error': '没有音频文件'}), 400
    
    file = request.files['audio']
    if file.filename == '':
        return jsonify({'error': '文件名为空'}), 400
    
    try:
        # 保存上传的音频
        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
        filename = f"recording_{timestamp}.webm"
        filepath = os.path.join(UPLOAD_FOLDER, filename)
        file.save(filepath)
        
        print(f"🎵 收到音频: {filename}, 大小: {os.path.getsize(filepath)} bytes")
        
        # 转写
        text, error = transcribe_audio(filepath)
        
        if error:
            print(f"❌ 转写失败: {error}")
            return jsonify({
                'success': False,
                'error': error,
                'filename': filename
            }), 500
        
        print(f"✅ 转写成功: {text[:50]}...")
        
        return jsonify({
            'success': True,
            'text': text or '（未识别到语音）',
            'filename': filename,
            'model_type': 'vosk' if os.path.exists(MODEL_PATH) else 'mock'
        })
        
    except Exception as e:
        print(f"❌ 服务器错误: {e}")
        return jsonify({
            'success': False,
            'error': str(e)
        }), 500


@app.route('/api/recordings', methods=['GET'])
def list_recordings():
    """列出所有录音文件"""
    recordings = []
    
    for filename in sorted(os.listdir(UPLOAD_FOLDER), reverse=True):
        if allowed_file(filename):
            filepath = os.path.join(UPLOAD_FOLDER, filename)
            stat = os.stat(filepath)
            recordings.append({
                'filename': filename,
                'size': stat.st_size,
                'size_human': f"{stat.st_size / 1024:.1f} KB",
                'created': datetime.fromtimestamp(stat.st_mtime).isoformat()
            })
    
    return jsonify({
        'count': len(recordings),
        'recordings': recordings
    })


@app.route('/api/recordings/<filename>', methods=['GET'])
def get_recording(filename):
    """获取单个录音文件"""
    return send_from_directory(UPLOAD_FOLDER, filename)


@app.route('/api/health', methods=['GET'])
def health_check():
    """健康检查"""
    model_exists = os.path.exists(MODEL_PATH)
    return jsonify({
        'status': 'ok',
        'model_loaded': model_exists,
        'model_path': MODEL_PATH,
        'upload_folder': UPLOAD_FOLDER,
        'recordings_count': len([f for f in os.listdir(UPLOAD_FOLDER) if allowed_file(f)])
    })


@app.route('/')
def index():
    return send_from_directory('../frontend', 'index.html')


@app.route('/<path:path>')
def static_files(path):
    return send_from_directory('../frontend', path)


if __name__ == '__main__':
    context = ssl.SSLContext(ssl.PROTOCOL_TLS_SERVER)
    context.load_cert_chain(
        '/home/ubuntu/nanocode/audio-recorder/cert.pem',
        '/home/ubuntu/nanocode/audio-recorder/key.pem'
    )
    
    print(f"🎙️ HTTPS 录音服务 V2 启动中...")
    print(f"📁 上传目录: {UPLOAD_FOLDER}")
    print(f"🔤 模型路径: {MODEL_PATH}")
    print(f"🔒 HTTPS 地址: https://129.211.173.87:8891")
    print(f"⚠️  提示: {'✅ 模型已安装' if os.path.exists(MODEL_PATH) else '⚠️ 使用模拟转写，请安装模型'}")
    
    app.run(
        host='0.0.0.0',
        port=8891,
        ssl_context=context,
        debug=False
    )