Ubuntu
107 lines
3.1 KiB
Python
Executable File
107 lines
3.1 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
"""
|
|
TTS 转换脚本 - 支持多语言
|
|
支持：中文(zh)、英文(en)、日文(jp)
|
|
"""
|
|
|
|
import argparse
|
|
import asyncio
|
|
import edge_tts
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
# 语音配置
|
|
VOICES = {
|
|
'zh': 'zh-CN-XiaoxiaoNeural', # 中文-晓晓
|
|
'zh-hk': 'zh-HK-HiuMaanNeural', # 粤语
|
|
'en': 'en-US-AriaNeural', # 英语-美音
|
|
'en-gb': 'en-GB-SoniaNeural', # 英语-英音
|
|
'jp': 'ja-JP-NanamiNeural', # 日语
|
|
'kr': 'ko-KR-SunHiNeural', # 韩语
|
|
}
|
|
|
|
# 默认语速
|
|
DEFAULT_RATE = "+0%"
|
|
|
|
|
|
async def convert_text_to_speech(text: str, voice: str, output_file: str, rate: str = DEFAULT_RATE):
|
|
"""将文本转换为语音"""
|
|
communicate = edge_tts.Communicate(text, voice, rate=rate)
|
|
await communicate.save(output_file)
|
|
print(f"✅ 已保存: {output_file}")
|
|
|
|
|
|
async def convert_file(input_file: str, lang: str = 'auto', output_file: str = None, rate: str = DEFAULT_RATE):
|
|
"""转换文件"""
|
|
input_path = Path(input_file)
|
|
|
|
if not input_path.exists():
|
|
print(f"❌ 文件不存在: {input_file}")
|
|
sys.exit(1)
|
|
|
|
# 读取文本
|
|
with open(input_file, 'r', encoding='utf-8') as f:
|
|
text = f.read()
|
|
|
|
# 自动检测语言（简单判断）
|
|
if lang == 'auto':
|
|
lang = detect_language(text)
|
|
print(f"🔍 检测到语言: {lang}")
|
|
|
|
# 获取语音
|
|
voice = VOICES.get(lang, VOICES['zh'])
|
|
|
|
# 确定输出文件名
|
|
if output_file is None:
|
|
output_file = str(input_path.with_suffix('.mp3'))
|
|
|
|
# 转换
|
|
print(f"🎙️ 使用语音: {voice}")
|
|
print(f"📝 转换: {input_file} -> {output_file}")
|
|
await convert_text_to_speech(text, voice, output_file, rate)
|
|
|
|
return output_file
|
|
|
|
|
|
def detect_language(text: str) -> str:
|
|
"""简单语言检测"""
|
|
# 日文假名
|
|
if any('\u3040' <= c <= '\u309f' or '\u30a0' <= c <= '\u30ff' for c in text[:100]):
|
|
return 'jp'
|
|
# 韩文
|
|
if any('\uac00' <= c <= '\ud7af' for c in text[:100]):
|
|
return 'kr'
|
|
# 中文
|
|
if any('\u4e00' <= c <= '\u9fff' for c in text[:100]):
|
|
return 'zh'
|
|
# 默认英文
|
|
return 'en'
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description='文本转语音工具')
|
|
parser.add_argument('input', help='输入文本文件')
|
|
parser.add_argument('--lang', '-l', default='auto',
|
|
choices=['auto', 'zh', 'zh-hk', 'en', 'en-gb', 'jp', 'kr'],
|
|
help='语言代码 (默认: auto)')
|
|
parser.add_argument('--output', '-o', help='输出音频文件')
|
|
parser.add_argument('--rate', '-r', default=DEFAULT_RATE,
|
|
help='语速调整，如 "+10%" 或 "-10%"')
|
|
parser.add_argument('--list-voices', action='store_true',
|
|
help='列出所有可用语音')
|
|
|
|
args = parser.parse_args()
|
|
|
|
if args.list_voices:
|
|
print("可用语音:")
|
|
for code, voice in VOICES.items():
|
|
print(f" {code}: {voice}")
|
|
return
|
|
|
|
# 运行转换
|
|
asyncio.run(convert_file(args.input, args.lang, args.output, args.rate))
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|