菜单
Clone a voice using qwen3-tts and generate speech from text
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Generate karaoke-style word-level timestamps by aligning script text to audio using Qwen3-ForcedAligner + jieba for Chinese word segmentation. Use when the user says 'align captions', 'karaoke timestamps', 'word timestamps', 'caption alignment', 'sync text to audio'.
Analyze raw video content using Gemini to identify speakers, topics, key moments, and potential clip opportunities
Audio processing utilities - noise reduction, normalization, enhancement
Extract a video segment using FFmpeg with precise start/end times
Find naturally clean, coherent video segments worth keeping (selection over repair)
Text-guided audio source separation using SAM-Audio via mlx-audio
| name | voice-clone |
| description | Clone a voice using qwen3-tts and generate speech from text |
Use this skill to clone a speaker's voice and generate text-to-speech audio.
python skills/voice-clone/clone.py <audio_sample.wav> [--transcript "text"]
Creates a speaker embedding file that can be reused.
python skills/voice-clone/speak.py <embedding.safetensors> "Text to speak"
Generates audio using the cloned voice.
assets/outputs/voice_embeddings/<name>_embedding.safetensors - Reusable voice modelassets/outputs/audio/<name>_speech.wav - Generated audio