# Transcript Kit

Transcribe your own audio or video and get captions. It is free, with no sign-up and no key.

## What it does

Turn a recording you own into a timestamped transcript with SRT and VTT captions, search what you have transcribed, find the best short clips in a talk, and translate the captions with a dub script.

Ask "transcribe this podcast episode", "make subtitles for my video", "where did I mention listed prices?", "find the best 60 seconds of this talk" or "translate these subtitles to Spanish".

Transcript Kit works on audio and video the user owns or has the rights to use: a file uploaded to the chat, a direct link to an audio or video file, or the link of a public video on TikTok, Instagram, Facebook, X, Vimeo or another social platform, whose audio is read without any account and deleted after the transcription. YouTube links go to get_transcript, which reads the captions. Each file can be up to 15 MB and 15 minutes long (mp3, m4a, mp4, wav, ogg, flac or webm).

Speech is transcribed with the Whisper large-v3-turbo model on Cloudflare Workers AI, which costs about 0.0005 dollars per audio minute. It is free to use, with no sign-up. To keep it free, each user and the service as a whole can transcribe only a limited number of minutes per day (UTC); every answer says how many minutes are left.

Each transcript is kept for 24 hours so it can be searched and clipped, then deleted, and a library holds the latest 10. The media file itself is never kept. The search finds passages and repeated themes in your own transcripts only. Clip suggestions are text with start and end times and a hook line; no video is cut.

Captions can be translated into about 100 languages with the m2m100 model on Cloudflare Workers AI (about 0.0002 dollars per thousand characters), keeping the original times, with a dub script of natural spoken lines sized to each segment. A call takes up to 18,000 characters, and translated text is not stored.

A transcript is a model's best reading of the audio and can contain mistakes, especially with names, accents or noise. Check it before you publish it.

- Transcribes an uploaded audio or video file or a direct file link, with timestamps
- Returns SRT and VTT caption files ready to use
- Accepts files up to 15 MB and 15 minutes in mp3, m4a, mp4, wav, ogg, flac or webm
- Searches your own transcripts for topics, hooks and repeated themes
- Suggests the best 30 to 60 second clips with start, end and a hook line, as text
- Translates captions into about 100 languages and writes a dub script sized to each segment
- Free with daily limits of minutes and characters; shows how much is left
- Keeps transcripts 24 hours, then deletes them; never keeps the media
- Transcribes a public TikTok, Instagram or other social video from its link, with no account

## Who it is for

Anyone who asks an AI assistant a question like this one: “Transcribe this recording and give me SRT captions.”

Roles that use it at work: Content creator.

## Price

Transcript Kit is free to use. There is no sign-up, no account and no API key. Calls are limited to 60 calls a minute per network and 5000 calls a day per server, shared by all callers. Repeat questions are answered from a cache.

## Tools it offers

- `transcribe_media`: Transcribe a recording into text and captions
- `search_transcripts`: Search your saved transcripts
- `suggest_clips`: Suggest short clips from a transcript
- `translate_subtitles`: Translate subtitles and make a dub script

## Connect

The MCP server is `https://transcript.openkrill.app/mcp` (streamable HTTP). It needs no key.

Claude Code, in one command:

```
claude mcp add --transport http transcript-kit https://transcript.openkrill.app/mcp
```

Hermes Agent, in one command that answers its two setup questions (no key, enable all tools):

```
printf 'n\ny\n' | hermes mcp add transcript-kit --url https://transcript.openkrill.app/mcp
```

Or add these lines to `~/.hermes/config.yaml`, under an existing `mcp_servers:` key if the file has one. Start a new Hermes session after either step.

```
mcp_servers:
  transcript-kit:
    url: https://transcript.openkrill.app/mcp
```

Any other agent: give it this page's Markdown address, `https://tools.openkrill.app/tools/transcript-kit.md`, and ask it to connect Transcript Kit.

Paste the address into your assistant once: the [connect steps](https://tools.openkrill.app/#connect) cover Claude, ChatGPT and Cursor, and the [install guides](https://tools.openkrill.app/integrations) show the same steps for each client.

## Job recipes for agents

Step-by-step recipes that walk an agent through 3 jobs with this server's tools:

- Transcribe a recording into text and captions, with `transcribe_media`
- Search your saved transcripts, with `search_transcripts`
- Translate subtitles and make a dub script, with `translate_subtitles`

[Download the recipes as SKILL.md](https://tools.openkrill.app/skills/transcript-kit-jobs/SKILL.md). Save it as `transcript-kit-jobs/SKILL.md` in your agent's skills folder. A connected MCP client gets the same recipes from `prompts/list` on the gateway, `https://api.openkrill.app/gateway/mcp`.

It is listed under Writing, media and files in the [full tool list](https://tools.openkrill.app/tools). The same facts in plain text are in [llms.txt](https://tools.openkrill.app/llms.txt).

Source: the Transcript Kit listing, version 1.0.1. As of 2026-10-18.