Why do I need to register words in different cases?

Many local TTS engines are case-sensitive. Registering 'Git', 'git', and 'GIT' ensures the engine recognizes the word regardless of how it appears in your text.

Which TTS engines is this skill compatible with?

This skill is designed for local Japanese TTS engines that utilize a compatible API structure, specifically COEIROINK and VOICEVOX.

Can I use this with OpenAI TTS?

OpenAI TTS handles English natively and usually doesn't require a Katakana dictionary. This skill is optimized for local engines that struggle with non-Japanese text.

Does this skill require an external API?

Yes, for the automatic reading generation feature (auto-add), you need a ZAI_API_KEY or OPENROUTER_API_KEY to leverage an LLM for Katakana conversion.

TTS Pronunciation Manager

Name: TTS Pronunciation Manager
Author: haboshi

byhaboshi

0•

生産性とワークフロー

Manages Text-to-Speech pronunciation dictionaries by automatically converting English terms into Katakana for local TTS engines.

The TTS Dictionary Manager is a specialized tool designed to bridge the gap between English terminology and local Japanese Text-to-Speech (TTS) engines like COEIROINK and VOICEVOX. It leverages LLMs to automatically generate accurate Katakana readings for technical terms, brand names, and jargon, ensuring smooth and natural audio output. By providing features for bulk scanning of scripts, case-variant registration, and direct engine synchronization, it eliminates the manual tedium of dictionary management and prevents the common 'robotic' mispronunciation of English words in generated speech.

主な機能

010 GitHub stars

02Case-variant support to register UPPERCASE, lowercase, and PascalCase versions of words

03Synchronization tools to apply dictionary updates directly to local TTS engines

04Bulk scanning of text and JSON files to extract and register new terms

05LLM-powered automatic Katakana reading generation for English vocabulary

06Integrated verification commands to audit actual engine pronunciation before final output

ユースケース

01Batch-processing dialogue files to prepare pronunciation dictionaries for local TTS narration

02Fixing specific mispronunciations in COEIROINK/VOICEVOX without manual GUI editing

03Ensuring technical terms like 'TypeScript' or 'GitHub' are pronounced correctly in automated voiceovers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add haboshi/claude-code-skills tts-dict

For use in Claude.ai and ChatGPT

Download Skill