File size: 4,371 Bytes
fdafed2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 |
---
tags:
- generated_from_trainer
datasets:
- evanarlian/common_voice_11_0_id_filtered
metrics:
- wer
model-index:
- name: wav2vec2-xls-r-164m-id
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: evanarlian/common_voice_11_0_id_filtered
type: evanarlian/common_voice_11_0_id_filtered
metrics:
- name: Wer
type: wer
value: 0.3199428097039019
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-xls-r-164m-id
This model is a fine-tuned version of [evanarlian/distil-wav2vec2-xls-r-164m-id](https://huggingface.co/evanarlian/distil-wav2vec2-xls-r-164m-id) on the evanarlian/common_voice_11_0_id_filtered dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3215
- Wer: 0.3199
## 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: 0.0002
- train_batch_size: 24
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.3
- num_epochs: 40.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 3.5445 | 0.92 | 1000 | 3.0106 | 1.0000 |
| 2.5067 | 1.84 | 2000 | 1.6134 | 0.9905 |
| 1.0279 | 2.75 | 3000 | 0.7667 | 0.8217 |
| 0.7823 | 3.67 | 4000 | 0.6141 | 0.7224 |
| 0.6504 | 4.59 | 5000 | 0.5228 | 0.6503 |
| 0.5687 | 5.51 | 6000 | 0.4666 | 0.5963 |
| 0.5026 | 6.43 | 7000 | 0.4288 | 0.5612 |
| 0.4584 | 7.35 | 8000 | 0.4048 | 0.5267 |
| 0.4193 | 8.26 | 9000 | 0.4057 | 0.5218 |
| 0.3931 | 9.18 | 10000 | 0.3820 | 0.4813 |
| 0.3651 | 10.1 | 11000 | 0.3686 | 0.4709 |
| 0.3526 | 11.02 | 12000 | 0.3665 | 0.4655 |
| 0.3333 | 11.94 | 13000 | 0.3440 | 0.4485 |
| 0.3095 | 12.86 | 14000 | 0.3314 | 0.4331 |
| 0.2802 | 13.77 | 15000 | 0.3360 | 0.4157 |
| 0.2724 | 14.69 | 16000 | 0.3331 | 0.4107 |
| 0.2488 | 15.61 | 17000 | 0.3255 | 0.4037 |
| 0.231 | 16.53 | 18000 | 0.3089 | 0.3950 |
| 0.2146 | 17.45 | 19000 | 0.3398 | 0.3990 |
| 0.2103 | 18.37 | 20000 | 0.3080 | 0.3805 |
| 0.2035 | 19.28 | 21000 | 0.3158 | 0.3828 |
| 0.1933 | 20.2 | 22000 | 0.3118 | 0.3728 |
| 0.1839 | 21.12 | 23000 | 0.3076 | 0.3690 |
| 0.1791 | 22.04 | 24000 | 0.3041 | 0.3658 |
| 0.1696 | 22.96 | 25000 | 0.3092 | 0.3603 |
| 0.1608 | 23.88 | 26000 | 0.2936 | 0.3555 |
| 0.1568 | 24.79 | 27000 | 0.2936 | 0.3560 |
| 0.1456 | 25.71 | 28000 | 0.3257 | 0.3543 |
| 0.1399 | 26.63 | 29000 | 0.3100 | 0.3424 |
| 0.1345 | 27.55 | 30000 | 0.3172 | 0.3472 |
| 0.1264 | 28.47 | 31000 | 0.3276 | 0.3412 |
| 0.1289 | 29.38 | 32000 | 0.3104 | 0.3401 |
| 0.1246 | 30.3 | 33000 | 0.3204 | 0.3352 |
| 0.1156 | 31.22 | 34000 | 0.3013 | 0.3353 |
| 0.1143 | 32.14 | 35000 | 0.3102 | 0.3322 |
| 0.1152 | 33.06 | 36000 | 0.3240 | 0.3323 |
| 0.1093 | 33.98 | 37000 | 0.3105 | 0.3295 |
| 0.101 | 34.89 | 38000 | 0.3112 | 0.3263 |
| 0.1017 | 35.81 | 39000 | 0.3263 | 0.3239 |
| 0.0915 | 36.73 | 40000 | 0.3176 | 0.3226 |
| 0.0943 | 37.65 | 41000 | 0.3141 | 0.3210 |
| 0.0898 | 38.57 | 42000 | 0.3177 | 0.3183 |
| 0.0923 | 39.49 | 43000 | 0.3215 | 0.3199 |
### Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2
|