whisper_DigitalUmuganda_Afrivoice_Fleurs_Shona_100hr_v2

This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2754
• Wer: 0.2605
• Cer: 0.0551

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.4453	1.0	4449	0.3022	0.2705	0.0545
0.2129	2.0	8898	0.2808	0.2575	0.0501
0.1421	3.0	13347	0.2915	0.2710	0.0522
0.0917	4.0	17796	0.3072	0.2613	0.0488
0.0547	5.0	22245	0.3334	0.2653	0.0509
0.032	6.0	26694	0.3684	0.2707	0.0480
0.0203	7.0	31143	0.3973	0.2619	0.0470
0.0146	8.0	35592	0.4232	0.2600	0.0495
0.0116	9.0	40041	0.4453	0.2599	0.0471
0.0095	10.0	44490	0.4572	0.2593	0.0490
0.0085	11.0	48939	0.4778	0.2595	0.0474
0.0075	12.0	53388	0.4916	0.2604	0.0474

Framework versions

• Transformers 4.45.2
• Pytorch 2.2.0+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1