Callmac Skill

Remote voice control for your Mac from mobile devices. Broadcast announcements, play alarms, tell stories, wake up kids - all triggered from messaging apps like Telegram or WhatsApp. Uses edge-tts with mixed Chinese/English support.

从移动设备远程控制 Mac 语音功能。广播公告、播放闹钟、讲故事、叫孩子起床 - 全部通过 Telegram 或 WhatsApp 触发。使用 edge-tts 支持中英文混合。

Quick Start | 快速开始

快速开始使用语音技能，生成和播放中英文混合的 TTS 音频。

Basic Usage | 基本用法

# Generate and play a simple announcement
python3 scripts/generate_tts.py --text "Hello world" --play

# Generate mixed Chinese/English content
python3 scripts/generate_tts.py --text "Hello 你好" --play

# Save to file
python3 scripts/generate_tts.py --text "Your message" --output announcement.mp3

Voice Selection

Edge TTS provides high-quality neural voices:

• English (US): en-US-JennyNeural (friendly), en-US-AriaNeural (confident)
• Chinese (Mandarin): zh-CN-XiaoxiaoNeural (warm), zh-CN-XiaoyiNeural (lively)
• Other languages: See VOICES.md for complete list

Features

1. Mixed Language Support

Automatically detects language segments and uses appropriate voices:

• English segments → English neural voice
• Chinese segments → Chinese neural voice
• Other languages → Default or specified voice

2. Playback Control

• Local playback on Mac using afplay
• Loop playback support (1-∞ times)
• Volume control (0-100%)
• Background/foreground playback options

3. File Management

• Save as MP3 files
• Concatenate multiple audio segments
• Batch processing support

4. Advanced Features

• Custom voice selection per segment
• Speech rate adjustment
• Pitch modification
• SSML support for advanced control

Workflows

Workflow 1: Simple Announcement Playback

1. User provides text to speak
2. System detects language(s)
3. Generates appropriate TTS audio
4. Plays locally on Mac

Workflow 2: Mixed Language Audio Creation

1. User provides Chinese/English mixed text
2. System splits by language segments
3. Generates separate audio for each language
4. Concatenates into single MP3 file
5. Optionally plays or saves

Workflow 3: Scheduled/Repeated Announcements

1. User provides text and playback schedule
2. System creates cron job or loop
3. Plays at specified intervals
4. Can be stopped on demand

Scripts

scripts/generate_tts.py

Main script for TTS generation and playback.

Usage:

# Basic generation and playback
python3 generate_tts.py --text "Message" --play

# Save to file
python3 generate_tts.py --text "Message" --output file.mp3

# Mixed language with custom voices
python3 generate_tts.py --text "Hello 你好" --voice-en "en-US-AriaNeural" --voice-zh "zh-CN-XiaoyiNeural"

# Loop playback
python3 generate_tts.py --text "Message" --play --loops 5

# Volume control
python3 generate_tts.py --text "Message" --play --volume 80

# SSML support
python3 generate_tts.py --ssml "<speak>Hello <break time='500ms'/> world</speak>"