DrishtiSharma/wav2vec2-large-xls-r-300m-bg-d2

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

wav2vec2-large-xls-r-300m-bg-d2

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMONVOICE8_0 - BG dataset. It achieves the following results on the evaluation set:

• —Loss: 0.3421
• —Wer: 0.2860

Evaluation Commands

1. 1.To evaluate on mozilla-foundation/commonvoice8_0 with test split

python eval.py --modelid DrishtiSharma/wav2vec2-large-xls-r-300m-bg-d2 --dataset mozilla-foundation/commonvoice80 --config bg --split test --log_outputs

1. 1.To evaluate on speech-recognition-community-v2/dev_data

python eval.py --modelid DrishtiSharma/wav2vec2-large-xls-r-300m-bg-d2 --dataset speech-recognition-community-v2/devdata --config bg --split validation --chunklengths 10 --stridelengths 1

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 0.00025
• —trainbatchsize: 16
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 32
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 700
• —num_epochs: 35
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.16.2
• —Pytorch 1.10.0+cu111
• —Datasets 1.18.3
• —Tokenizers 0.11.0