Publisher: OpenAI · Editorial profile by AI Tools Ratings · Review method
Not runtime tested; source-document review only. No security certification.
What task is this skill designed for?
Transcribe a short synthetic interview and check names, timestamps and speaker changes before using the output in a research note.
What input and output should you expect?
Input: An audio file you are authorized to process, desired output format, a language hint and optional speaker-reference clips when supported.
Output: Expected, not independently tested: a transcript file or structured transcript with available speaker segments. Speaker labels describe segments and are not a verified identity determination.
What does it need to run?
Python, the OpenAI client dependency, an appropriately configured API key and access to a transcription model. The pinned source’s model choices and limits need checking against current API documentation before a real request.
Authored for Codex in the OpenAI skills repository. This directory has not independently installed or executed the skill.
How should you approach installation?
- Inspect the pinned source and helper, then check current transcription API requirements.
- Configure the API key locally and use a short non-sensitive audio sample.
- Review transcript accuracy and processing cost before authorizing a larger batch.
Installation guide: discovery, dependencies and runtime checks
What files, accounts or data can it access?
Audio is submitted to the configured API service. This differs from a local-only transcription engine. Recordings can contain personal or business information; confirm authority to process them and keep the API key in the local secret environment.
What does the skill cost?
API transcription is a separate usage-based service; the skill does not make it free or include API credit. Check current service pricing before submitting recordings.
What are its limits and alternatives?
Noise, overlapping speech, specialist terms and short speaker turns can produce errors. Automatic labels do not establish who a person is. Do not quote critical statements without checking the relevant audio segment.
Use an approved local transcription engine when audio must remain on device. Choose this workflow when the API processing route is acceptable and a repeatable CLI output is useful.
What could you ask in a small trial?
Try this with a small sample:
Transcribe this authorized synthetic interview to text with timestamps if supported. Flag uncertain names, save the result locally, and do not infer speaker identities from voices.
How would you verify the result?
- Compare several transcript passages with the original audio, including names and numbers.
- Check speaker-boundary changes and mark uncertainty.
- Confirm output location, API processing route and actual request cost before scaling to many recordings.
Use a baseline and record acceptance evidence before treating a successful task as a reliable workflow.
Which source and license were checked?
Read the reviewed SKILL.md at its pinned commit · Complete reviewed folder
Repository and current publisher guidance
Checked:
Apache-2.0 — identified in skills/.curated/transcribe/LICENSE.txt at the reviewed commit. Third-party assets and dependencies may have separate terms. License evidence
Version details and review scope
Source: openai/skills/skills/.curated/transcribe
Reviewed commit: 49f948faa9258a0c61caceaf225e179651397431
SKILL.md SHA-256: f530021da9f377362ee4214e0bc689cbb5404640d2eb88ee272d51d61f530a09
Source-document review. Source text and stated dependencies were inspected for this profile. Referenced code was not comprehensively audited, no runtime test was performed, and source files are not mirrored here.