abulkhaera

EmotionVoice — AI skill for Claude Code

AI community

Local web app + CLI for emotional AI text-to-speech and voice cloning with the Fish Audio API.

How to install EmotionVoice

This entry records only its repository, not the path inside it, so there is no exact command to give. Open abulkhaera/EmotionVoice and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What EmotionVoice does

Local web app + CLI for emotional AI text-to-speech and voice cloning with the Fish Audio API. 60+ emotion tags, 6 languages (EN/ID/ZH/JA/KO/AR), one-click Windows launcher, ready-made Claude skill to use it with your Claude.

Alternatives in AI

Om Humanize — Voice-calibrated editing — makes Claude-drafted text sound like you wrote it, not like AI wrote it 4.6k ★
Qwen Audio Agent — A realtime voice runtime that keeps Agents talking, working, and present 2.3k ★
Claude Skill Web Clone — Methodology for faithfully cloning any website (static / React / WebGL) — without copying AI-hallucinated code 969 ★

README

EmotionVoice

Turn text into **emotional, expressive speech** — and clone your own voice — with the [Fish Audio](https://fish.audio) API. A small local web app plus a CLI. One double-click to start on Windows.

[excited] We finally did it!
[sighing] It took so long though.
[whispering] Don't tell anyone yet.

↑ Type that, press **Generate**, hear three different emotions in one clip.

Features

60+ emotion & delivery tags — happy, sad, sarcastic, nostalgic, whispering, shouting, laughing, sighing, pauses… One click inserts a tag at the cursor. S2 models also accept free-form cues like [slightly nervous but hopeful].
6 languages, native voices — English, Bahasa Indonesia, Mandarin, Japanese, Korean, Arabic. Picking a language filters the voice library to native speakers; Arabic switches the editor to right-to-left.
Clone your own voice — upload clips or record straight from the mic (converted to WAV in the browser). Voices are saved private in your Fish Audio account and ready instantly.
Voice library search — browse public voices or only your own.
Fine control — model, format (mp3 / wav / opus), speed, volume, expressiveness, diversity.
History & auto-save — every take plays in the page and is saved to output/.
CLI for scripts and AI agents — voice_cli.py returns clean JSON. Includes a ready-made Claude skill (SKILL.md).
Key stays private — your API key lives in .env on the local server and is never sent to the browser.

Quick start (Windows)

Install Python 3.10+ from python.org — tick "Add python.exe to PATH".
Get an API key at fish.audio/app/api-keys.
Double-click start.bat. On first run it creates .env and opens it in Notepad.
Paste your key after FISH_API_KEY=, save, close.
Double-click start.bat again. It sets up a virtual env, installs packages, and opens **http://127