video-voice-changer

Extract audio from video files, apply voice conversion, and merge it back with FFmpeg.

611|75|Updated Jan 10, 2026
One-click install
npx skills add https://github.com/wlzh/skills --skill video-voice-changer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-voice-changer
Source: https://github.com/wlzh/skills/tree/main/video-voice-changer
Command: npx skills add https://github.com/wlzh/skills --skill video-voice-changer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Changing the voice in a video normally requires manual audio extraction, separate voice processing, and re-muxing with video editing tools. This Skill automates that entire pipeline so a video's audio track can be voice-converted in a single command while keeping the original video quality untouched.

Core Features & Use Cases

  • Automated Pipeline: Extracts audio with FFmpeg, delegates voice conversion to the voice-changer skill, and re-merges the result without re-encoding the video stream.
  • Multiple Voice Presets: Supports presets such as female_1, female_2, male_deep, male_normal, child, and robot via the shared voice-changer configuration file.
  • Safe Output Behavior: Creates a new file with a _vc suffix by default and only overwrites the original when --overwrite is passed; optional --keep-audio saves the converted audio copy.
  • Use Case: A content creator wants to publish a screen recording with a different voice for privacy. Running the script on the MP4 produces a new video with a converted voice track in under a minute.

Quick Start

Ask the AI to convert the voice in your video file to a deep male voice and save it as a new file without overwriting the original.

Frequently Asked Questions about video-voice-changer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I change the voice in a video file?

Run the video_voice_change.py script with your video path, for example python3 video_voice_change.py input.mp4 -v male_deep. It extracts the audio with FFmpeg, converts the voice, and merges it back into a new video file.

What voice presets are available for video voice conversion?

Available presets include female_1, female_2, female_3, male_deep, male_normal, child, and robot. They are loaded from the voice-changer skill's voice_config.json, so custom presets added there can also be used.

Does video voice conversion reduce video quality?

No. The video stream is copied without re-encoding using FFmpeg's -c:v copy flag, so the original video quality is preserved. Only the audio track is replaced and encoded as AAC.

Does the tool overwrite the original video file?

By default it creates a new file with a _vc suffix, such as input_vc.mp4, leaving the original untouched. The original is only overwritten when you explicitly pass the --overwrite flag.

Why does video voice conversion fail with a dependency error?

The script requires FFmpeg, FFprobe, Python 3.8+, and the voice-changer skill installed at ~/.claude/skills/voice-changer. Install FFmpeg via brew or apt and verify the voice-changer script and config exist.

What are the limitations of FFmpeg-based video voice changing?

Processing time scales with video length, roughly 30-60 seconds for a 10-minute 1080p video. It also needs temporary disk space for extracted audio, and output audio quality depends on the voice-changer skill's conversion quality.