whisper-small-se / README.md
TeoJM's picture
Create README.md
81b26c3
|
raw
history blame
853 Bytes
metadata
language: sv
license: apache-2.0
tags:
  - seq-2-seq asr
  - whisper
datasets:
  - mozilla-foundation/common_voice_11_0
library_name: torch

Model description

This model is a fine-tuned version of OpenAI's pre-trained Whisper small. Source

The model were trained on 680 000 hours of audio with corresponding transcripts from the internet, 65% of which was in english audio and 83 % of which had english transcripts.

The model was then further trained for 4000 iterations, 500 of which as warm-up, on Swedish data from Common_voice 11.0. Achieving a WER of 19.865.

Training Metrics

Training table

Model Plot

View Training Plots

Metrics