ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v2

gary109/ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v2

ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v2 is a automatic speech recognition model from gary109. Use it when you need speech turned into text. It is set up for transformers.

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

Downloads · 30 days

2% of all-time downloads

All-time downloads

883

Public

Repo size

1.5 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin378 MB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
Model type: wav2vec2
Access: Public
Created: Nov 22, 2022
Updated: Nov 22, 2022
SHA: d64c5b3f

Datasets

ai_light_dance

Model card

From the Hugging Face model README

ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v2

This model is a fine-tuned version of gary109/ai-light-dance_drums_pretrain_wav2vec2-base-new on the GARY109/AI_LIGHT_DANCE - ONSET-IDMT-SMT-DRUMS-V2+MDBDRUMS dataset. It achieves the following results on the evaluation set:

Loss: 0.5264
Wer: 0.3635

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: 0.0004
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 10
num_epochs: 100.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
4.6468	0.98	22	3.2315	1.0
1.5745	1.98	44	3.1603	1.0
1.465	2.98	66	2.2551	1.0
1.3168	3.98	88	1.8461	1.0
1.1359	4.98	110	1.4874	0.9797
0.9769	5.98	132	1.7359	0.5495
0.9019	6.98	154	1.5833	0.5268
0.8057	7.98	176	1.4892	0.5304
1.0845	8.98	198	1.3939	0.5197
0.7562	9.98	220	1.1238	0.5447
0.7259	10.98	242	1.2936	0.5006
0.7318	11.98	264	1.2763	0.4660
0.6452	12.98	286	1.2947	0.4779
0.6353	13.98	308	1.1925	0.4517
0.6463	14.98	330	0.8667	0.4100
0.5381	15.98	352	1.1243	0.3909
0.5637	16.98	374	0.8683	0.3754
0.6149	17.98	396	1.1040	0.3731
0.6138	18.98	418	1.1068	0.3850
0.7381	19.98	440	0.9203	0.3623
0.5064	20.98	462	0.8806	0.3540
0.4731	21.98	484	0.7259	0.3623
0.5232	22.98	506	0.7935	0.3516
0.4689	23.98	528	0.7771	0.3540
0.4902	24.98	550	0.6897	0.3909
0.4079	25.98	572	0.8030	0.3552
0.5045	26.98	594	0.6778	0.3790
0.4373	27.98	616	0.7456	0.3695
0.4366	28.98	638	0.7009	0.3433
0.3944	29.98	660	0.6841	0.3468
0.4206	30.98	682	0.7093	0.3373
0.3949	31.98	704	0.6901	0.3576
0.4416	32.98	726	0.6762	0.3397
0.4248	33.98	748	0.7196	0.3540
0.4214	34.98	770	0.6669	0.3254
0.416	35.98	792	0.6422	0.3445
0.3687	36.98	814	0.6345	0.3504
0.4119	37.98	836	0.6306	0.3385
0.359	38.98	858	0.6538	0.3576
0.359	39.98	880	0.6613	0.3349
0.3488	40.98	902	0.5976	0.3468
0.3543	41.98	924	0.6327	0.3433
0.3647	42.98	946	0.6208	0.3600
0.3529	43.98	968	0.6008	0.3492
0.3691	44.98	990	0.6065	0.3492
0.329	45.98	1012	0.6288	0.3373
0.3357	46.98	1034	0.5760	0.3480
0.3318	47.98	1056	0.5637	0.3564
0.3181	48.98	1078	0.5560	0.3468
0.3313	49.98	1100	0.5905	0.3337
0.3059	50.98	1122	0.5443	0.3278
0.3375	51.98	1144	0.5695	0.3576
0.3191	52.98	1166	0.5874	0.3385
0.3115	53.98	1188	0.5264	0.3635
0.3044	54.98	1210	0.5480	0.3433
0.3256	55.98	1232	0.5677	0.3385
0.2938	56.98	1254	0.5597	0.3445
0.2853	57.98	1276	0.5942	0.3373
0.3348	58.98	1298	0.5733	0.3421
0.3024	59.98	1320	0.5604	0.3433
0.2655	60.98	1342	0.5348	0.3468
0.3029	61.98	1364	0.5752	0.3206
0.3435	62.98	1386	0.5489	0.3063
0.3125	63.98	1408	0.5736	0.3075
0.263	64.98	1430	0.5505	0.3206
0.2665	65.98	1452	0.5391	0.3230
0.299	66.98	1474	0.5389	0.3135
0.2909	67.98	1496	0.5841	0.3099
0.2988	68.98	1518	0.5847	0.3004
0.2879	69.98	1540	0.5941	0.2968
0.2802	70.98	1562	0.6612	0.2920
0.2877	71.98	1584	0.5641	0.3051
0.2727	72.98	1606	0.6138	0.3063
0.2668	73.98	1628	0.6087	0.2920
0.2675	74.98	1650	0.5876	0.2932
0.264	75.98	1672	0.6043	0.2980
0.2352	76.98	1694	0.5829	0.2932
0.2494	77.98	1716	0.5775	0.3063
0.2621	78.98	1738	0.5676	0.2956
0.2788	79.98	1760	0.5864	0.2932
0.2615	80.98	1782	0.5754	0.3015
0.2542	81.98	1804	0.5651	0.3027
0.2641	82.98	1826	0.5731	0.3004
0.2532	83.98	1848	0.5782	0.2968
0.2645	84.98	1870	0.5718	0.3039
0.2296	85.98	1892	0.5628	0.3147
0.2394	86.98	1914	0.5920	0.3027
0.2636	87.98	1936	0.6085	0.2968
0.2371	88.98	1958	0.5809	0.3075
0.2364	89.98	1980	0.5927	0.3039
0.2812	90.98	2002	0.5713	0.3123
0.2141	91.98	2024	0.5743	0.3039
0.2919	92.98	2046	0.5837	0.3063
0.2288	93.98	2068	0.5860	0.3015
0.2585	94.98	2090	0.5776	0.3147
0.2529	95.98	2112	0.5625	0.3159
0.2343	96.98	2134	0.5700	0.3087
0.2567	97.98	2156	0.5729	0.3087
0.2448	98.98	2178	0.5728	0.3111
0.2501	99.98	2200	0.5744	0.3099

Framework versions

Transformers 4.25.0.dev0
Pytorch 1.8.1+cu111
Datasets 2.7.1.dev0
Tokenizers 0.13.2

Task: Automatic Speech Recognition
Library: transformers
Type: wav2vec2
Created: Nov 22, 2022
Updated: Nov 22, 2022