wav2vec2-large-xls-r-300m-assamese_speech_to_IPA_with_wer_cer_f1

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8170
• Exact Match: 0.0325
• F1: 0.0325
• Wer: 0.9675
• Cer: 0.1450

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0003
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Exact Match	F1	Wer	Cer
4.6801	9.8765	400	0.8184	0.0032	0.0032	0.9968	0.2199
0.2678	19.7531	800	0.7753	0.0227	0.0227	0.9773	0.1628
0.1009	29.6296	1200	0.8270	0.0292	0.0292	0.9708	0.1504
0.0619	39.5062	1600	0.8170	0.0325	0.0325	0.9675	0.1450

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1