About
The fish-audio skill enables AI-powered audio generation through Fish Audio models, providing text-to-speech conversion in multiple languages and audio transcription capabilities. Developers can use it to create high-quality voice narration from text input or transcribe existing audio files to text. It supports customizable voice styles and handles both file generation and text extraction workflows.
Quick Install
Claude Code
Recommendednpx skills add refly-ai/refly-skills -a claude-code/plugin add https://github.com/refly-ai/refly-skillsgit clone https://github.com/refly-ai/refly-skills.git ~/.claude/skills/fish-audioCopy and paste this command in Claude Code to install this skill
GitHub Repository
Frequently asked questions
What is the fish-audio skill?
fish-audio is a Claude Skill by refly-ai. Skills package instructions and resources that Claude loads on demand, so Claude can perform fish-audio-related tasks without extra prompting.
How do I install fish-audio?
Use the install commands on this page: add fish-audio to Claude Code as a plugin, or clone its repository into your skills directory, then restart Claude so it picks up the skill.
What category does fish-audio belong to?
fish-audio is in the file-generation category, tagged ai.
Is fish-audio free to use?
Yes. fish-audio is listed on AIMCP and free to install. It runs inside Claude, so no separate service account is required to use the skill itself.
Related Skills
This skill generates AI videos via ByteDance's Seedance models through the Volcengine Ark API. It supports text-to-video, image-to-video (using first/last frames or reference images), and task management. Developers must configure an `ARK_API_KEY` to use models like Seedance 1.5 Pro (with audio) and 1.0 variants.
This Claude Skill generates AI videos via the BytePlus Seedance API, allowing developers to create videos from text prompts, images, or by managing existing tasks. It supports multiple models including Seedance 1.5 Pro with audio and draft mode capabilities. Use it when you need to programmatically produce or query AI-generated video content within your applications.
The `wan-video` skill enables AI video generation using Alibaba's Wan 2.6 models. Developers can use it to create videos from text prompts, animate static images, or apply AI style transfer to existing videos. It supports multiple generation modes and accepts JSON input with file references for processing.
The seedream-image skill generates AI images using ByteDance's Seedream 4.5 model, primarily creating visuals from text prompts. It supports style transfer transformations and produces high-quality artistic or realistic images in various aspect ratios. Developers should use it when needing image generation, editing, or batch creation capabilities within their applications.