Which OpenAI models are used for speech generation?

The skill defaults to 'gpt-4o-mini-tts-2025-12-15' but also supports standard 'tts-1' and 'tts-1-hd' models.

Can I create custom voices with this skill?

No, custom voice creation is out of scope. The skill uses OpenAI's high-quality built-in voices such as Cedar and Marin.

Does this skill support batch processing?

Yes, the skill can handle multiple lines or prompts at once using a batch workflow and temporary JSONL files for efficiency.

Do I need an API key to use this skill?

Yes, an active OpenAI API key must be configured in your environment as 'OPENAI_API_KEY' for the skill to make live API calls.

What is the maximum text length for a single request?

The input length is limited to 4096 characters per request, though longer texts can be split into multiple chunks or batch processed.

OpenAI Speech Synthesis

Name: OpenAI Speech Synthesis
Author: x-cmd

byx-cmd

•

생산성 및 워크플로

Generates high-quality AI narration and voiceovers for projects using the OpenAI Audio API and specialized delivery controls.

The Speech Generation skill enables Claude to produce professional-grade audio narration, product demo voiceovers, and accessibility reads directly within the development environment. By leveraging the OpenAI Audio API and a dedicated CLI script, it supports both single-clip generation and high-volume batch processing. Users can fine-tune output through instruction augmentation—adjusting voice affect, tone, pacing, and emphasis—to ensure generated audio meets specific requirements for IVR prompts, educational content, or UI feedback.