config.py

"""
CosyVoice Configuration
配置CosyVoice模型的参数和设置
"""

import os

# 模型配置
MODEL_CONFIG = {
    # 模型路径配置
    "model_dir": os.environ.get("COSYVOICE_MODEL_DIR", "pretrained_models/CosyVoice-300M"),
    
    # Hugging Face模型名称
    "hf_model_name": "FunAudioLLM/CosyVoice-300M",
    
    # 备用模型名称
    "alternative_models": [
        "FunAudioLLM/CosyVoice-300M-SFT",
        "FunAudioLLM/CosyVoice-300M-Instruct",
    ],
    
    # 设备配置
    "device": "cuda" if os.environ.get("CUDA_VISIBLE_DEVICES") else "cpu",
    
    # 音频配置
    "sample_rate": 22050,
    "max_duration": 30,  # 最大音频时长（秒）
}

# 说话人配置
SPEAKER_CONFIG = {
    "default_speaker": "中文女",
    "available_speakers": [
        "中文女",
        "中文男",
        "英文女",
        "英文男",
        "粤语女",
        "粤语男",
        "日语男",
        "韩语女",
    ],
}

# Gradio界面配置
GRADIO_CONFIG = {
    "title": "CosyVoice - 语音处理模型",
    "description": """
CosyVoice是一个先进的语音处理模型，支持语音识别、语音合成等功能。

上传音频文件或使用麦克风录制语音，体验CosyVoice的强大功能。
    """,
    "theme": "soft",
    "share": True,
}

# 部署配置
DEPLOYMENT_CONFIG = {
    # Hugging Face Space配置
    "hf_space": {
        "sdk": "gradio",
        "sdk_version": "6.4.0",
        "app_file": "app.py",
        "emoji": "🌍",
        "colorFrom": "blue",
        "colorTo": "pink",
        "license": "apache-2.0",
    },
    
    # 环境变量
    "env_vars": {
        "GRADIO_SERVER_NAME": "0.0.0.0",
        "GRADIO_SERVER_PORT": "7860",
    },
}

# 日志配置
LOGGING_CONFIG = {
    "level": "INFO",
    "format": "%(asctime)s - %(name)s - %(levelname)s - %(message)s",
}