Skip to the card

Card 944 of 9932026-09-21 issue

Observed arrival · 2026-09-21

Voxint — Local Speaker-Labeled Transcripts

voxint.app Visit website
Editorial interest 84/100 Selection signal · not a rating of the site

An open-source tool that turns recordings into speaker-labelled transcripts on your own hardware.

Landing page captured for the 2026-09-21 issue.
For
Researchers, journalists, educators, and small teams
Worth noticing
Speaker proposals remain separate from user rulings, creating a review queue and provenance trail.

Field notes

Voxint separates the automated transcript and speaker proposals from the operator’s corrections, making review part of the workflow rather than an afterthought. Its installation path uses Docker Compose and a provided shell script, while the site says processing can run on CPU with roughly 8 GB of free memory. The project identifies Whisper, pyannote, and TitaNet as its local processing components and offers several text, subtitle, and structured-data exports.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

OpenPublic substance visible
PrettyNotable craft visible
ProPolished or operationally mature
NicheUnusually specific use

One card from the complete issue

A Passport-Sized Place to Begin

246,713 arrived 999 judged 993 catalogued Enter the complete issue
voxint.app

Landing page observed 2026-09-21. The live site may have changed.