Audio Transcriber

Productivity skill, available on Zeplik

Audio Transcriber is a ready-to-run productivity skill on Zeplik. Not for summarizing documents (use summarize-anything). Ask in plain language and Zeplik applies the skill's method for you inside the conversation, on whichever AI model you prefer.

The Audio Transcriber skill loads automatically when your request matches it, or you can invoke it directly by typing /transcribe in any chat. It works with attachments, connectors, and any model that supports the task, so you get the same expert method every time without setting anything up.

What the Audio Transcriber skill can do

Try these prompts on Zeplik

Pick a prompt to open it in the Zeplik app. If you are not signed in yet, your prompt is waiting for you the moment you do.

How the Audio Transcriber skill works

/transcribe

Turn attached audio into useful text with Hermes-native voice transcription. The transcription itself is the easy part; the value is in what you do with it: cleanup, speaker labeling, timestamps, and shaping the raw transcript into the artifact the user actually needs.

For condensing documents or already-written text, this is the wrong skill -- use summarize-anything.

Workflow

  1. Get the audio. Ask the user to attach the recording if they have not. Note the context they give (meeting, interview, lecture, voice memo) and who the speakers are.
  2. Transcribe the attachment with Hermes-native transcription.
  3. Ask (or infer) the target artifact -- verbatim transcript, cleaned transcript, minutes, notes, or quotes -- and produce it per the playbooks below.

Output playbooks

Verbatim transcript

Faithful to the audio, including false starts and fillers only if the user asks (legal, research, and journalism contexts often need them). Mark unintelligible spans as [inaudible], uncertain words as [word?]. Never silently guess.

Cleaned transcript (default)

  • Remove fillers (um, uh, you know), false starts, and stutters.
  • Fix punctuation, casing, and paragraph breaks at topic shifts.
  • Preserve meaning and voice exactly; cleanup is not paraphrase.

Speaker labeling

  • Use names when the user supplies them or when speakers identify themselves in the audio; otherwise Speaker 1, Speaker 2.
  • Label every turn: **Alice:** .... When attribution is uncertain, mark it **Speaker 1(?):** rather than guessing.
  • Offer the user a quick mapping step: "Speaker 2 sounds like the interviewer -- who was that?"

Timestamps

  • Add [mm:ss] markers at speaker turns or topic changes (user's choice; default to topic changes for long recordings).
  • For quotes the user will cite, always attach the timestamp.

Minutes / notes from the recording

Structure:

## Meeting Minutes -- [topic, date if known]
**Attendees:** [speakers identified]
**Summary:** [2-3 sentences]

### Decisions
- [decision] ([who], [timestamp])

### Action Items
- [ ] [action] -- [owner] -- [due if stated]

### Discussion Notes
[Topic-by-topic bullets with timestamps]

### Open Questions
- [unresolved items]

Only record decisions and actions actually stated in the audio; do not infer commitments nobody made.

Quality standards

  • Flag audio quality problems (crosstalk, low volume, heavy accent segments) instead of papering over them.
  • Keep numbers, names, and technical terms exact; mark them [?] when unsure.
  • For multi-language audio, transcribe in the spoken language and offer a translation as a separate artifact.
  • Long recordings: deliver in sections and confirm direction after the first section.

Usage

/transcribe $ARGUMENTS

How to use the Audio Transcriber skill

  1. Sign in to Zeplik

    Create a free Zeplik account or sign in. New accounts start with free credits, so you can try the Audio Transcriber skill right away.

  2. Describe your productivity task

    Ask in plain language, or type /transcribe to invoke the skill directly. Zeplik recognizes the Audio Transcriber skill and applies its method.

  3. Review and refine the result

    Zeplik returns a clear, structured answer. Ask follow-ups in the same chat to refine it or take the next step.

Source and credit

Author
davila7
License
MIT

Adapted from the open-source davila7/claude-code-templates project and tuned to run natively on Zeplik. View source on GitHub.

Frequently asked questions

What is the Audio Transcriber skill?
Audio Transcriber is a ready-to-run productivity skill on Zeplik. Not for summarizing documents (use summarize-anything). Ask in plain language and Zeplik applies the skill's method for you inside the conversation, on whichever AI model you prefer.
How do I use Audio Transcriber on Zeplik?
Sign in to Zeplik and ask in plain language, or type /transcribe in any chat to invoke it directly. The skill applies its method and returns a result you can refine in the same conversation.
Which AI model does the Audio Transcriber skill use?
Any model you choose. Zeplik works across every model in one chat, so the Audio Transcriber skill runs on your preferred model for the task.
Where does the Audio Transcriber skill come from?
The Audio Transcriber skill is adapted from the open-source davila7/claude-code-templates project (MIT) and tuned to run natively on Zeplik. The original source is linked on this page.
How much does the Audio Transcriber skill cost?
Using the skill is free to start. You only spend Zeplik credits when the assistant runs, and new accounts begin with free credits.

Related productivity skills

More on Zeplik

Try Audio Transcriber on Zeplik

Every model, one chat. Bring the Audio Transcriber skill into your next conversation and let the assistant do the work.

Browse all skills
Audio Transcriber - Productivity skill for Zeplik AI | Zeplik Chat