Easy-to-use benchmarks using audio and video content from the Internet Archive, specifically targeting various challenging scenarios in audio recordings.
Find a file
2023-12-02 04:31:39 +00:00
LICENSE Initial commit 2023-12-02 03:16:17 +00:00
README.md Update README.md 2023-12-02 04:31:39 +00:00

WhisperBenchmarks

This repository provides easy-to-use benchmarks using audio and video content from the Internet Archive, specifically targeting various challenging scenarios in audio recordings.

Categories Title Links
Poor mic placement Body camera footage from July 10 traffic stop Internet Archive
Thick accents Moonshine for Medicine Popcorn Sutton Internet Archive
Low-quality audio 1994 90210 Melrose Place Promos Commercial Internet Archive
Artifacts in audio 2002 007 Movie Trailer Commercial Bad Video Internet Archive
Ideal audio (one speaker) 8 Bit Bookclub Internet Archive

How to Run Whisper Benchmarks

-- TODO --

Results

Links are embeded for each category

CPU Benchmarks

CPU Model Poor mic placement (s) Thick accents (s) Low-quality audio (s) Artifacts in audio (s) Ideal audio (one speaker) (Docker/Native)

GPU Benchmarks

GPU Model Poor mic placement (s) Thick accents (s) Low-quality audio (s) Artifacts in audio (s) Ideal audio (one speaker) (Docker/Native)

Example

Todo:

  • Write easy bash scripts for running a set of benchmarks with an easy cleanup
  • Finalize a standard format for exporting the data into a spreadsheet