Transcription Service Demo

Upload a WAV, MP3, MP4, or MOV file to test the transcription engine

Test the Transcription Service

The example below uses the OpenAI Whisper model to transcribe a myriad of hospital video to be accessibility compliant. This model one runs solely on my own machine so there's no additional licensing or usage costs, but please be patient as the model can take a couple of minutes to process.

File and bandwidth limits: 100MB and 10 minutes

This is a Flask Python implementation and hosted with Nginx.

Select a media file to upload. Supported formats: MP3, WAV, MP4, MOV.

Sample Files

English WAV Sample

An 11-second audio clip from John F. Kennedy's speech.

jfk.wav

Japanese WAV Sample

A 10-second audio clip from Shogun in Japanese.

Shogun_jp.wav

Multi-Voice MP3 Sample

A 35-second audio clip from Conan The Barbarian with multiple speakers and heavy accents

Conan_What_is_best_in_life.mp3

Home Video Sample

A 33-second video clip of me talking to my cat.

cat_negotiation.MOV