Transform transcribe audio with transloadit

Skill Pyfagorass/bookofspells/skills/transloadit/transform-transcribe-audio-with-transloadit

📖 The Book of Spells: a curated, enchanted index of real LLM tooling — and a pipeline that gathers SKILL.md skills from many houses into one searchable shelf.

Install

npx -y skills add Pyfagorass/bookofspells --skill transform-transcribe-audio-with-transloadit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

One-off transcription of local audio or video files to text or subtitle files using Transloadit via the official `@transloadit/node` CLI. Use when the user wants speech in local media converted to `.txt`, `.json`, `.srt`, or `.webvtt`; prefer the `speech transcribe` intent with Replicate while ElevenLabs Scribe is not ready.

SKILL.md

2.8 KB, 595 tokens by cl100k_base, as published. Nobody here has run it

Transcribe Audio or Video with Transloadit

Use this for

Local audio or video files that should become text transcripts or subtitle files
One or many files, with outputs downloaded locally
Agent workflows where credentials may already be in the shell, a local .env, or ~/.transloadit/credentials

Inputs

Absolute path to one or more local audio or video files
Optional output path; default is the same basename with .txt
Optional format: text, json, srt, or webvtt

Workflow

Confirm each input file exists and resolve it to an absolute path.
Derive the output path:
- Single input and no requested output: same directory and basename with .txt
- Multiple inputs and no requested output: a transcripts/ directory next to the inputs
- Use .json, .srt, or .webvtt when the user asks for that format
Let the CLI resolve auth automatically in this order:
- Shell environment variables
- The current working directory .env only
- ~/.transloadit/credentials If your .env lives in a parent directory, export the variables into the shell first.
Run the transcription with the official CLI intent:

npx -y @transloadit/node speech transcribe --input /ABS/PATH/input.opus --output /ABS/PATH/input.txt

For multiple files, repeat --input and write to an output directory:

npx -y @transloadit/node speech transcribe \
  --input /ABS/PATH/a.opus \
  --input /ABS/PATH/b.opus \
  --output /ABS/PATH/transcripts/

Options

Default format: text
Other formats: --format json, --format srt, --format webvtt
Default provider: replicate until ElevenLabs Scribe is ready
Supported explicit providers: --provider aws, --provider gcp, --provider replicate
Do not use --provider transloadit; that internal Whisper route is work in progress
Use --source-language <code> when the input language is known
Use --target-language <code> when translation is requested

Notes

Prefer @transloadit/node; it is the official CLI route and exposes speech transcribe.
If the command errors with a missing subcommand, update to a newer @transloadit/node release that includes the speech transcription intent.
Keep secrets local-only; never print TRANSLOADIT_SECRET or copy it into browser code.
After transcription, confirm each expected output file exists and is non-empty.