openclip-transcription

Transcribe video and audio files into JSON, SRT, and VTT with speaker diarization.

1|Updated Aug 7, 2026
One-click install
npx skills add https://github.com/OpenClip-App/agent-skills --skill openclip-transcription-openclip-app
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openclip-transcription
Source: https://github.com/OpenClip-App/agent-skills/tree/main/skills/openclip-transcription
Command: npx skills add https://github.com/OpenClip-App/agent-skills --skill openclip-transcription-openclip-app

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually transcribing video or audio recordings is slow and error-prone, and producing properly time-coded subtitle files in multiple formats adds even more work. This Skill automates the full pipeline from raw media upload to finished transcript. ## Core Features & Use Cases - Multi-format transcription: Convert any uploaded video or audio file into structured JSON (with segment timings), SRT, and VTT outputs hosted on permanent CDN URLs. - Speaker diarization: Automatically label who is speaking, with the option to disable it for clean single-speaker output. - Language control: Auto-detect the language or set it explicitly to avoid misdetection on short or noisy clips. - Use Case: You have a recorded meeting and need to know who said what. Upload the file, run transcription with diarization enabled, poll the job status, and read the JSON output grouped by speaker with exact time codes. ## Quick Start Ask your agent to transcribe the attached interview video and return the transcript with speaker labels and an SRT subtitle file.

Frequently Asked Questions about openclip-transcription

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I transcribe a video file to text?▼

Upload the file with create_upload, PUT the raw bytes to the returned upload_url, then call transcribe with the upload id. Poll get_tool_job_status every 5 to 10 seconds until the job completes and returns JSON, SRT, and VTT output URLs.

How to generate an SRT subtitle file from a video?▼

Run the transcribe tool on your uploaded video and wait for the job to complete. The completed job returns permanent CDN output URLs, one per format, so you can hand the user the srt URL directly as a ready subtitle file.

Does transcription support speaker diarization?▼

Yes, the transcribe tool labels who is speaking by default through the diarize parameter. Turn it off for single-speaker audio to keep the output clean, or leave it on to group JSON segments by speaker.

Can I transcribe audio in a specific language like Dutch?▼

Yes, pass the language parameter to transcribe, for example language="nl" for Dutch. Omitting it triggers auto-detection, but setting it explicitly prevents misdetection on short or noisy clips.

What is the difference between transcribe and get_transcript?▼

transcribe is the free tool that works on any freshly uploaded file. get_transcript reads transcripts from videos that already went through the paid clipping pipeline and supports start_ms and end_ms windows at sentence or word level.

Why does transcription not return text immediately?▼

Transcription is asynchronous and never returns text synchronously. The transcribe call returns a tool_job id, and you must poll get_tool_job_status until it reaches completed, failed, or returns an error.