You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
154 lines
4.0 KiB
154 lines
4.0 KiB
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
"""
|
|
实时语音转文字系统主程序
|
|
基于sherpa-onnx的实时语音识别应用
|
|
"""
|
|
|
|
import sys
|
|
import argparse
|
|
import logging
|
|
from pathlib import Path
|
|
|
|
# 添加src目录到Python路径
|
|
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
|
|
|
from src import RealTimeVTT, ModelDownloader, ModelConfig
|
|
|
|
def setup_logging(level: str = "INFO"):
|
|
"""设置日志"""
|
|
logging.basicConfig(
|
|
level=getattr(logging, level.upper()),
|
|
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
|
|
)
|
|
|
|
def check_models() -> bool:
|
|
"""检查模型文件是否存在"""
|
|
config = ModelConfig()
|
|
missing_files = config.validate_model_files()
|
|
|
|
if missing_files:
|
|
print("错误: 缺少模型文件")
|
|
print("缺少的文件:")
|
|
for file_path in missing_files:
|
|
print(f" - {file_path}")
|
|
print("\n请运行以下命令下载模型:")
|
|
print(" python main.py --download-model")
|
|
return False
|
|
|
|
return True
|
|
|
|
def download_model_interactive():
|
|
"""交互式下载模型"""
|
|
config = ModelConfig()
|
|
downloader = ModelDownloader(config)
|
|
downloader.interactive_download()
|
|
|
|
def main():
|
|
"""主函数"""
|
|
parser = argparse.ArgumentParser(
|
|
description="实时语音转文字系统",
|
|
formatter_class=argparse.RawDescriptionHelpFormatter,
|
|
epilog="""
|
|
示例用法:
|
|
python main.py # 启动实时语音识别
|
|
python main.py --download-model # 下载语音识别模型
|
|
python main.py --list-devices # 列出音频设备
|
|
python main.py --log-level DEBUG # 启用调试日志
|
|
"""
|
|
)
|
|
|
|
parser.add_argument(
|
|
"--download-model",
|
|
action="store_true",
|
|
help="下载语音识别模型"
|
|
)
|
|
|
|
parser.add_argument(
|
|
"--list-devices",
|
|
action="store_true",
|
|
help="列出可用的音频设备"
|
|
)
|
|
|
|
parser.add_argument(
|
|
"--log-level",
|
|
choices=["DEBUG", "INFO", "WARNING", "ERROR"],
|
|
default="INFO",
|
|
help="设置日志级别 (默认: INFO)"
|
|
)
|
|
|
|
parser.add_argument(
|
|
"--no-save",
|
|
action="store_true",
|
|
help="不保存识别结果到文件"
|
|
)
|
|
|
|
parser.add_argument(
|
|
"--no-partial",
|
|
action="store_true",
|
|
help="不显示部分识别结果"
|
|
)
|
|
|
|
args = parser.parse_args()
|
|
|
|
# 设置日志
|
|
setup_logging(args.log_level)
|
|
|
|
try:
|
|
if args.download_model:
|
|
# 下载模型
|
|
download_model_interactive()
|
|
return
|
|
|
|
# 创建应用实例
|
|
app = RealTimeVTT()
|
|
|
|
# 应用配置
|
|
if args.no_save:
|
|
app.app_config.save_to_file = False
|
|
|
|
if args.no_partial:
|
|
app.app_config.show_partial_results = False
|
|
|
|
if args.list_devices:
|
|
# 列出音频设备
|
|
if not app.audio_processor.initialize():
|
|
print("错误: 无法初始化音频设备")
|
|
return 1
|
|
|
|
devices = app.list_audio_devices()
|
|
print("可用的音频设备:")
|
|
for device in devices:
|
|
print(f" [{device['index']}] {device['name']}")
|
|
print(f" 通道数: {device['channels']}")
|
|
print(f" 采样率: {device['sample_rate']} Hz")
|
|
|
|
app.cleanup()
|
|
return
|
|
|
|
# 检查模型文件
|
|
if not check_models():
|
|
return 1
|
|
|
|
# 初始化应用
|
|
if not app.initialize():
|
|
print("错误: 应用初始化失败")
|
|
return 1
|
|
|
|
# 运行应用
|
|
app.run_interactive()
|
|
|
|
except KeyboardInterrupt:
|
|
print("\n程序被用户中断")
|
|
except Exception as e:
|
|
print(f"错误: {e}")
|
|
if args.log_level == "DEBUG":
|
|
import traceback
|
|
traceback.print_exc()
|
|
return 1
|
|
|
|
return 0
|
|
|
|
if __name__ == "__main__":
|
|
sys.exit(main())
|