wav2vec2-base-vietnamese-VIVOS-CommonVoice-FOSD-Control-dataset-25e-epochs

tuanmanh28/wav2vec2-base-vietnamese-VIVOS-CommonVoice-FOSD-Control-dataset-25e-epochs

wav2vec2-base-vietnamese-VIVOS-CommonVoice-FOSD-Control-dataset-25e-epochs is a automatic speech recognition model from tuanmanh28. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as cc-by-nc-4.0.

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

Downloads · 30 days

27% of all-time downloads

All-time downloads

Public

Repo size

24.9 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
License: cc-by-nc-4.0
Model type: wav2vec2
Access: Public
Created: Jan 17, 2024
Updated: Jan 18, 2024
SHA: ef5dc6c0

Base models

nguyenvulebinh/wav2vec2-base-vi

Model card

From the Hugging Face model README

wav2vec2-base-vietnamese-VIVOS-CommonVoice-FOSD-Control-dataset-25e-epochs

This model is a fine-tuned version of nguyenvulebinh/wav2vec2-base-vi on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3338
Wer: 0.1833

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: 32
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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
16.1039	0.39	500	21.1164	1.0
10.5383	0.77	1000	15.8037	1.0
7.5435	1.16	1500	9.8785	1.0
5.1426	1.55	2000	5.9691	1.0
3.9112	1.93	2500	4.1400	1.0
3.5159	2.32	3000	3.6877	1.0
3.4056	2.71	3500	3.5166	1.0
3.384	3.09	4000	3.6170	1.0
3.3715	3.48	4500	3.5045	1.0
3.373	3.87	5000	3.4859	1.0
3.3539	4.25	5500	3.4843	1.0
3.3063	4.64	6000	3.3596	1.0
3.0749	5.03	6500	2.8515	0.9994
2.6888	5.41	7000	2.4817	1.0000
2.3404	5.8	7500	2.0490	0.9815
2.0588	6.19	8000	1.7986	0.9288
1.8428	6.57	8500	1.4945	0.8332
1.686	6.96	9000	1.3796	0.7640
1.5399	7.35	9500	1.2362	0.6927
1.4374	7.73	10000	1.1130	0.6320
1.3281	8.12	10500	1.0058	0.5705
1.2308	8.51	11000	0.8888	0.5109
1.1405	8.89	11500	0.8438	0.4524
1.0647	9.28	12000	0.7767	0.4208
1.0104	9.67	12500	0.7385	0.3777
0.9629	10.05	13000	0.6731	0.3505
0.9045	10.44	13500	0.6295	0.3317
0.8573	10.83	14000	0.6071	0.3115
0.8443	11.21	14500	0.5895	0.2984
0.7915	11.6	15000	0.5828	0.2823
0.7965	11.99	15500	0.5552	0.2714
0.7738	12.37	16000	0.5100	0.2605
0.7326	12.76	16500	0.4884	0.2499
0.7007	13.15	17000	0.4799	0.2402
0.6997	13.53	17500	0.4647	0.2331
0.68	13.92	18000	0.4469	0.2271
0.6707	14.31	18500	0.4261	0.2231
0.6557	14.69	19000	0.4145	0.2164
0.6509	15.08	19500	0.4010	0.2120
0.6649	15.47	20000	0.4038	0.2092
0.6191	15.85	20500	0.3926	0.2064
0.6385	16.24	21000	0.3882	0.2024
0.6222	16.63	21500	0.3874	0.2016
0.5792	17.01	22000	0.3873	0.2023
0.5775	17.4	22500	0.3757	0.1975
0.5647	17.79	23000	0.3626	0.1964
0.5723	18.17	23500	0.3574	0.1958
0.5573	18.56	24000	0.3530	0.1960
0.5813	18.95	24500	0.3541	0.1933
0.563	19.33	25000	0.3455	0.1926
0.5402	19.72	25500	0.3483	0.1910
0.5578	20.11	26000	0.3516	0.1915
0.5456	20.49	26500	0.3477	0.1878
0.5453	20.88	27000	0.3391	0.1882
0.5265	21.27	27500	0.3386	0.1869
0.557	21.66	28000	0.3388	0.1864
0.5526	22.04	28500	0.3373	0.1864
0.5284	22.43	29000	0.3352	0.1854
0.5351	22.82	29500	0.3373	0.1850
0.5775	23.2	30000	0.3382	0.1848
0.5292	23.59	30500	0.3371	0.1843
0.52	23.98	31000	0.3338	0.1839
0.5372	24.36	31500	0.3337	0.1829
0.5167	24.75	32000	0.3338	0.1833

Framework versions

Transformers 4.32.1
Pytorch 2.0.1+cu117
Datasets 2.14.4
Tokenizers 0.13.3

Task: Automatic Speech Recognition
Library: transformers
Type: wav2vec2
License: cc-by-nc-4.0
Created: Jan 17, 2024
Updated: Jan 18, 2024