Skip to main content
Glama

Transcribe media from a URL

transcribe_media

Converts audio from video, podcast, or social media URLs into timestamped text; uses existing captions first, then AI speech-to-text when captions are missing.

Instructions

Transcribe the audio of any video/audio URL (YouTube, TikTok, podcasts, and 1000+ sites) to text with timestamped segments. One call usually returns the finished transcript; long media returns a jobId to poll with check_transcription. By default (mode "auto") it uses the video's own captions when it has them ($0.003) and AI speech-to-text when it has none ($0.01/min). Set maxCostDollars to cap the speech-to-text charge on long media; mode "captions" never runs speech-to-text. Runs the same job as POST /transcribe with mode "auto" and wait in the REST API; to build transcription into the user's app, call get_integration_guide.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "url": "<url>"
    },
    "name": "transcribe_media"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlYesMedia page URL, e.g. https://www.youtube.com/watch?v=...
modeNoauto (default): captions, then AI speech-to-text if there are none. captions: captions only ($0.003), fails with a cost quote if there are none. ai: always AI speech-to-text ($0.01/min)
languageNoISO language hint, e.g. "en"
providerNoSame as mode "ai": always use AI speech-to-text, even when the media has captions ($0.01/min)
maxCostDollarsNoAbort with a quote if the estimated speech-to-text cost exceeds this cap

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.2.1
    • changedInput schema / properties / maxCostDollars / description
      Previous value: -"Abort with a quote if the estimated cost exceeds this cap"New value: +"Abort with a quote if the estimated speech-to-text cost exceeds this cap"
    • addedInput schema / properties / mode
      Added value: +{
      +  "description": "auto (default): captions, then AI speech-to-text if there are none. captions: captions only ($0.003), fails with a cost quote if there are none. ai: always AI speech-to-text ($0.01/min)",
      +  "enum": [
      +    "auto",
      +    "captions",
      +    "ai"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / provider
      Added value: +{
      +  "description": "Same as mode \"ai\": always use AI speech-to-text, even when the media has captions ($0.01/min)",
      +  "enum": [
      +    "importly"
      +  ],
      +  "type": "string"
      +}
  2. First observedv0.1.0

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare openWorldHint=true, readOnlyHint=false, idempotentHint=false, so the agent knows this reaches external sites and creates a job. The description adds substantial context beyond that: per-mode pricing ($0.003 captions, $0.01/min AI), that mode "captions" never triggers speech-to-text, and that maxCostDollars aborts with a quote. It stops short of describing rate limits or expected latency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action and scope, then return behavior, cost model, and alternatives in descending priority. Dense but each sentence carries information; the REST endpoint parity note (POST /transcribe with mode "auto" and wait) is the least essential line.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 5-parameter, no-output-schema tool, this covers what an agent needs: what comes back (timestamped segments or jobId), how to continue (check_transcription), cost exposure, and the integration path. Nothing material about calling it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description earns a bit more by tying maxCostDollars to long-media cost control and explaining the default auto path. Much of the mode/cost wording duplicates the schema's own enum descriptions, limiting the added value.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Transcribe the audio of any video/audio URL") plus scope (YouTube, TikTok, podcasts, 1000+ sites) and output shape (timestamped segments). It also explicitly separates itself from check_transcription, so an agent can route without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete when-to-use guidance: one call usually returns the transcript, long media returns a jobId to poll with check_transcription, and app builders should call get_integration_guide instead. Mode selection is framed by cost, though it does not state failure/edge cases for unsupported URLs.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.