Skip to main content
Glama

Transcribe Speech

transcribe
Destructive

Convert speech from video or audio files into SRT subtitles locally. Generate subtitle files for captioning without cloud services.

Instructions

Transcribe a recording's speech to an SRT subtitle file, locally, with FFmpeg 8's whisper filter (needs a whisper.cpp model file). Returns the SRT path and the text. Pair with burn_subtitles to caption the video.

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": {
      "input": "<input>"
    },
    "name": "transcribe"
  }
}

Input Schema

NameRequiredDescriptionDefault
inputYesPath to the video or audio to transcribe.
modelNoPath to a whisper.cpp model file (ggml-*.bin). Defaults to SCREENCAST_WHISPER_MODEL.
wordsNoOne word per cue, for word timing (cut_by_transcript).
outputNoOptional output .srt path. Defaults under SCREENCAST_HOME/edits.
languageNoSpoken language code, e.g. 'en', or 'auto' (default).
overwriteNoAllow replacing an existing file at the output path (default false).

Output Schema

NameRequiredDescriptionDefault
cuesYes
textYes
srtPathYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.47.0
    • addedInput schema / properties / words
      Added value: +{
      +  "description": "One word per cue, for word timing (cut_by_transcript).",
      +  "type": "boolean"
      +}
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "$schema": "http://json-schema.org/draft-07/schema#",
      +  "additionalProperties": {},
      +  "properties": {
      +    "cues": {
      +      "maximum": 9007199254740991,
      +      "minimum": -9007199254740991,
      +      "type": "integer"
      +    },
      +    "srtPath": {
      +      "type": "string"
      +    },
      +    "text": {
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "srtPath",
      +    "cues",
      +    "text"
      +  ],
      +  "type": "object"
      +}
  2. First observedv0.32.0

TDQS

A4.2/5.0
Behavior4/5

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

Annotations already declare destructiveHint=true and readOnlyHint=false; the description adds useful context that processing is local, that an external whisper.cpp model file is required, and what is produced (SRT path plus text). The destructiveness is implied by writing an SRT file (mitigated by overwrite=false default documented in schema) but never explicitly discussed, leaving a small gap.

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

Conciseness5/5

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

Three compact sentences: the action and output first, the engine/prerequisite second, and the sibling pairing last. No filler, every clause carries information.

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

Completeness4/5

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

With an output schema present and annotations covering the mutation profile, the description needn't restate return values, and it still gives a one-line summary of them. It doesn't cover failure modes (missing model, unsupported language) or rough runtime expectations, which for a moderately complex local inference tool would be the finishing touch.

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

Parameters3/5

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

Schema coverage is 100%, so all six parameters (input, model, words, output, language, overwrite) are already documented. The description reinforces the model requirement but adds no syntax, format, or trade-off detail beyond the schema, so the baseline 3 applies.

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 (transcribe), resource (a recording's speech), and concrete output artifact (an SRT subtitle file), plus names the engine used. An agent can immediately distinguish this from siblings like burn_subtitles or cut_by_transcript, both of which are explicitly referenced or implied.

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 clear context ('Pair with burn_subtitles to caption the video') and a prerequisite ('needs a whisper.cpp model file'), which routes the agent correctly. It stops short of explicit when-not-to-use guidance or naming an alternative transcription path, so it's a strong 4 rather than a 5.

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