ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1

gary109/ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1

ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1 is a automatic speech recognition model from gary109. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

8% of all-time downloads

All-time downloads

598

Public

Repo size

25.2 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.3 GB · 99%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: wav2vec2
Access: Public
Created: Jul 8, 2022
Updated: Jul 19, 2022
SHA: c8a4dda3

Model card

From the Hugging Face model README

ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1

This model is a fine-tuned version of gary109/ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1 on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING3 dataset. It achieves the following results on the evaluation set:

Loss: 0.5459
Wer: 0.2463

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-06
train_batch_size: 2
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: 1000
num_epochs: 50.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3909	1.0	2309	0.5615	0.2459
0.4094	2.0	4618	0.5654	0.2439
0.326	3.0	6927	0.5568	0.2470
0.4577	4.0	9236	0.5795	0.2474
0.3628	5.0	11545	0.5459	0.2463
0.3135	6.0	13854	0.5582	0.2473
0.5058	7.0	16163	0.5677	0.2439
0.3188	8.0	18472	0.5646	0.2445
0.3589	9.0	20781	0.5626	0.2479
0.4021	10.0	23090	0.5722	0.2452
0.4362	11.0	25399	0.5659	0.2431
0.3215	12.0	27708	0.5658	0.2445
0.3646	13.0	30017	0.5785	0.2459
0.3757	14.0	32326	0.5757	0.2418
0.3311	15.0	34635	0.5672	0.2455
0.3709	16.0	36944	0.5669	0.2434
0.3342	17.0	39253	0.5610	0.2455
0.3236	18.0	41562	0.5652	0.2436
0.3566	19.0	43871	0.5773	0.2407
0.2912	20.0	46180	0.5764	0.2453
0.3652	21.0	48489	0.5732	0.2423
0.3785	22.0	50798	0.5696	0.2423
0.3968	23.0	53107	0.5690	0.2429
0.2968	24.0	55416	0.5800	0.2427
0.428	25.0	57725	0.5704	0.2441
0.383	26.0	60034	0.5739	0.2450
0.3694	27.0	62343	0.5791	0.2437
0.3449	28.0	64652	0.5780	0.2451
0.3008	29.0	66961	0.5749	0.2418
0.3939	30.0	69270	0.5737	0.2424
0.3451	31.0	71579	0.5805	0.2402
0.3513	32.0	73888	0.5670	0.2379
0.3866	33.0	76197	0.5706	0.2389
0.3831	34.0	78506	0.5635	0.2401
0.3641	35.0	80815	0.5708	0.2405
0.3345	36.0	83124	0.5699	0.2405
0.2902	37.0	85433	0.5711	0.2373
0.2868	38.0	87742	0.5713	0.2389
0.3232	39.0	90051	0.5702	0.2392
0.3277	40.0	92360	0.5658	0.2393
0.3234	41.0	94669	0.5732	0.2412
0.3625	42.0	96978	0.5740	0.2396
0.4075	43.0	99287	0.5733	0.2389
0.3473	44.0	101596	0.5735	0.2394
0.3157	45.0	103905	0.5721	0.2391
0.3866	46.0	106214	0.5715	0.2381
0.4062	47.0	108523	0.5711	0.2380
0.3871	48.0	110832	0.5716	0.2380
0.2924	49.0	113141	0.5723	0.2374
0.3655	50.0	115450	0.5709	0.2379

Framework versions

Transformers 4.21.0.dev0
Pytorch 1.9.1+cu102
Datasets 2.3.3.dev0
Tokenizers 0.12.1

Task: Automatic Speech Recognition
Library: transformers
Type: wav2vec2
License: apache-2.0
Created: Jul 8, 2022
Updated: Jul 19, 2022