swahili_v2 – AI Model by etonkou | AlphaNeural AI

You can deploy this model and start earning money today!

etonkou/
swahili_v2

like

0

automatic-speech-recognition

generated_from_trainer

facebook/wav2vec2-base

endpoints_compatible

Views

No views yet

swahili_v2

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5957
Wer: 0.4820

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: 0.0001
train_batch_size: 3
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.0806	0.6	1000	1.2990	0.9279
1.3262	1.2	2000	0.9359	0.8004
1.0778	1.8	3000	1.0139	0.7421
0.9227	2.4	4000	0.7815	0.7053
0.8283	3.0	5000	0.6969	0.6340
0.683	3.6	6000	0.6665	0.6254
0.6208	4.2	7000	0.6304	0.5900
0.561	4.8	8000	0.5912	0.5748
0.4881	5.4	9000	0.5998	0.5523
0.474	6.0	10000	0.5488	0.5409
0.3989	6.6	11000	0.5550	0.5274
0.3581	7.2	12000	0.5702	0.5088
0.3348	7.8	13000	0.5739	0.5034
0.3059	8.4	14000	0.5915	0.5042
0.2765	9.0	15000	0.5720	0.4860
0.2489	9.6	16000	0.5957	0.4820

Framework versions

Transformers 4.33.3
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3