YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Fine-tuning of wav2vec2-base
on 100h of Librispeech training data. Results on "clean" data are very similar to the ones of the official model. However, the result on "other" is significantly worse - the model seems to have overfitting to the "clean" data.
Model was trained on librispeech-clean-train.100 with following hyper-parameters:
- 2 GPUs Titan RTX
- Total update steps 13000
- Batch size per GPU: 32 corresponding to a total batch size of ca. ~1500 seconds
- Adam with linear decaying learning rate with 3000 warmup steps
- dynamic grouping for batch
- fp16
- attention_mask was not used during training
Result (WER) on Librispeech test:
"clean" | "other" |
---|---|
6.5 | 18.7 |
- Downloads last month
- 2