marma's picture
Update README.md
15aff64
|
raw
history blame
No virus
676 Bytes
metadata
license: apache-2.0
datasets:
  - KBLab/rixvox
language:
  - sv

Whisper Large RixVox Swedish

This is a Whisper large finetuned for Swedish using the RixVox dataset.

Evaluation

Common Voice 11:

  • WER: XYZ
  • WER (normalized): XYZ

Training

Training was done using Huggingface and Deepspeed with ZeRO stage 2.

  • learning rate: 1e-5
  • optimizer: CPUAdamW (Deepspeed)
  • lr scheduler: linear
  • warmup steps: 500
  • per device batch size: 20
  • GPUs: 8 x NVIDIA A100 40GB
  • total batch size: 160
  • steps: 12000
  • lowercase: no
  • fp16
  • entire encoder was frozen