edgespeak-align

Installation
SKILL.md

EdgeSpeak Align

Force-align an audio/video file against a reference transcript you already have, producing word-level timestamps — when each word starts and ends. Runs entirely on-device; the audio never leaves the machine. Under the hood it calls edgespeak-cli align. When the EdgeSpeak desktop app is running, the CLI talks to its local gateway (OpenAI-compatible, 127.0.0.1:1117) and reuses the warm model (proxy mode); when the app is not running, the CLI launches the bundled on-device engine itself (standalone mode). Standalone is a normal mode, not an error.

Version compatibility. The frontmatter pins this skill's version and the oldest CLI it is written against (minCliVersion). If edgespeak-cli --version reports something older, run edgespeak-cli update (or re-run the installer) before relying on the flags documented here. Same-numbered builds can still differ, so --help is the tiebreaker: a command or flag documented here but missing from the installed --help also means update — don't route around it.

Alignment ≠ transcription. Transcription guesses the words; alignment is given the words and only finds the timing. If the user does not have the text yet, use edgespeak-transcribe instead.

Inputs to confirm

  • Media path to align.
  • Reference transcript/script/lyrics text.
  • Desired output: stdout text, .txt, .json, or .srt.
  • Optional protected terms for brand names, jargon, names, or tokens that must stay verbatim.

How to do it

  1. Confirm two inputs: the media file and the reference text (a string, or a text file to read).
  2. Check the runtime first:
Installs
5
GitHub Stars
6
First Seen
Jul 28, 2026
edgespeak-align — lattifai/edgespeak-skills