Back to catalog / Developer tools
OpenAI Speech: generate narration from the exact text
Use the bundled CLI with the OpenAI Audio API for single or JSONL batch jobs.
Skills
Collect the exact text, built-in voice, delivery style, format, and constraints first. For multiple lines, write a temporary JSONL file and run the bundled `scripts/text_to_speech.py` once. Each request is limited to 4,096 characters and the CLI caps throughput at 50 RPM; state the default model, voice, and flags used. Live calls require the user to set `OPENAI_API_KEY` in the environment; never ask them to paste it in chat. If needed, install the `openai` dependency following project rules, but do not modify the bundled CLI. For important clips, check intelligibility, pacing, pronunciation, and constraints. Save output under a stable `output/speech/` path and disclose that audio is AI-generated.
Permissions
Prerequisites
- For text-to-speech, narration, accessibility reads, or batch audio generation.
- Specify exact text, built-in voice, delivery, format, single/batch mode, and output path; live calls need `OPENAI_API_KEY`.
Permissions
- May create temporary JSONL and call a paid OpenAI API to generate audio; outputs go to the chosen path.
- Use built-in voices only, disclose AI generation, and never ask for an API key in chat.
Limitations
Use the skill CLI with `--out` or `--out-dir`; each request is limited to 4096 characters and 50 requests/minute. Do not modify the bundled CLI.
Pinned upstream source
- Repository
- openai/skills
- File path
- skills/.curated/speech/SKILL.md
- Pinned commit
49f948faa9258a0c61caceaf225e179651397431- Source checked
- 2026-10-14 · Reviewed against the pinned upstream source
Repository metrics
These values describe the upstream repository, not popularity or likes for an individual skill or playbook.
- Stars
- Not collected
- Forks
- Not collected
- Observed at
- Not collected
No metrics snapshot is available yet.
License: Apache-2.0
Compatible tools
bundled `scripts/text_to_speech.py` · OpenAI Python SDK · built-in TTS voices
Main topic
Take a coding Agent change from context to verification