360 Secure SkillHub
360 安全技能中心

Pywayne Tts

基于 pywayne.tts 的文本转语音工具,提供 text_to_speech_output_opus 与 text_to_speech_output_mp3 两个函数,把输入文本合成为 Opus 或 MP3 音频文件。macOS 上默认调用系统 say 命令,其他平台使用 Google TTS 服务,并借助 ffmpeg 完成音频格式转换与临时文件清理。

wangyendt
wangyendt
热度 74.00
最近更新 2026-05-11 17:25:56
分类:创作依赖:代码执行 / 读写文件 / 第三方 API应用场景:文件转换 / 工作流

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add tts-2
Metadata
name
pywayne-tts
description
Text-to-speech conversion tool. Use when converting text to speech audio files (opus or mp3 format). Supports macOS native 'say' command and Google TTS (gTTS) service with ffmpeg audio conversion.

Pywayne TTS

Text-to-speech (TTS) tool that converts text to audio files (opus or mp3 format).

Functions

text_to_speech_output_opus - Convert text to Opus format

from pywayne.tts import text_to_speech_output_opus

# Output: test.opus
text_to_speech_output_opus("你好,世界", "test.opus")

Parameters:

  • text - Text to convert
  • opus_filename - Output .opus filename
  • use_say - If True, use macOS 'say' command; otherwise use gTTS (default True)

Behavior:

  • On macOS, prefers native 'say' command when use_say=True
  • On other platforms, uses Google TTS (gTTS) service
  • Uses ffmpeg to convert audio to opus format (16kHz, mono channel)
  • Automatically cleans up temporary files

text_to_speech_output_mp3 - Convert text to MP3 format

from pywayne.tts import text_to_speech_output_mp3

# Output: test.mp3
text_to_speech_output_mp3("你好,世界", "test.mp3")

Parameters:

  • text - Text to convert
  • mp3_filename - Output .mp3 filename
  • use_say - If True, use macOS 'say' command; otherwise use gTTS (default True)

Behavior:

  • On macOS, prefers native 'say' command when use_say=True
  • On other platforms, uses Google TTS (gTTS) service
  • Uses ffmpeg to convert audio to mp3 format
  • Automatically cleans up temporary files

Quick Start

# Convert text to Opus format (default: macOS uses 'say')
text_to_speech_output_opus "你好,世界" "test.opus"

# Convert text to MP3 format
text_to_speech_output_mp3 "你好,世界" "test.mp3"

# Force use gTTS instead of macOS 'say'
text_to_speech_output_mp3 "你好,世界" "test.mp3" use_say=False

Requirements

  • ffmpeg: Required for audio conversion
    • macOS: brew install ffmpeg
    • Windows: Download from https://ffmpeg.org and add to PATH
    • Linux: sudo apt install ffmpeg or package manager
  • gtts: Python library for Google TTS service

Platform Detection

Module automatically detects platform and prompts for ffmpeg installation if missing.

Audio Formats

  • Opus: Better quality, smaller file size, suitable for voice calls
  • MP3: Better compatibility, suitable for multimedia playback

Notes

  • Requires network connection for gTTS service
  • Temporary files are automatically cleaned up
  • Ensure output directory has write permissions