asr-amharic-phoneme-based-37-5

Samuael/asr-amharic-phoneme-based-37-5

asr-amharic-phoneme-based-37-5 is a automatic speech recognition model from Samuael. 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

19% of all-time downloads

All-time downloads

Public

Parameters

315M

3.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1.3 GB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: wav2vec2
Access: Public
Created: Jan 15, 2024
Updated: Jan 18, 2024
SHA: ce97c52d

Model card

From the Hugging Face model README

asr-amharic-phoneme-based-37-5

This model is a fine-tuned version of Samuael/asr-amharic-phoneme-based-37-5 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2362
Wer: 0.3159
Phoneme Cer: 0.0591
Cer: 0.0858

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: 3e-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: 100
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Phoneme Cer	Cer
0.0861	0.59	200	0.3128	0.3362	0.0633	0.0924
0.1616	1.18	400	0.2772	0.3354	0.0626	0.0910
0.1407	1.76	600	0.2752	0.3362	0.0631	0.0921
0.1267	2.35	800	0.2659	0.3367	0.0629	0.0912
0.1042	2.94	1000	0.2607	0.3338	0.0623	0.0905
0.1487	3.53	1200	0.2612	0.3255	0.0614	0.0899
0.09	4.12	1400	0.2679	0.3297	0.0623	0.0899
0.1001	4.71	1600	0.2655	0.3239	0.0607	0.0883
0.0863	5.29	1800	0.2723	0.3234	0.0608	0.0881
0.1198	5.88	2000	0.2530	0.3273	0.0608	0.0885
0.23	6.47	2200	0.2576	0.3273	0.0613	0.0892
0.1518	7.06	2400	0.2565	0.3191	0.0604	0.0874
0.2198	7.65	2600	0.2539	0.3134	0.0590	0.0858
0.1592	8.24	2800	0.2499	0.3174	0.0595	0.0863
0.1395	8.82	3000	0.2423	0.3190	0.0598	0.0865
0.163	9.41	3200	0.2401	0.3171	0.0590	0.0858
0.1252	10.0	3400	0.2362	0.3159	0.0591	0.0858

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0

Task: Automatic Speech Recognition
Library: transformers
Type: wav2vec2
License: apache-2.0
Created: Jan 15, 2024
Updated: Jan 18, 2024