Commit
·
fdafed2
1
Parent(s):
cf558e9
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,114 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- evanarlian/common_voice_11_0_id_filtered
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: wav2vec2-xls-r-164m-id
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Automatic Speech Recognition
|
| 13 |
+
type: automatic-speech-recognition
|
| 14 |
+
dataset:
|
| 15 |
+
name: evanarlian/common_voice_11_0_id_filtered
|
| 16 |
+
type: evanarlian/common_voice_11_0_id_filtered
|
| 17 |
+
metrics:
|
| 18 |
+
- name: Wer
|
| 19 |
+
type: wer
|
| 20 |
+
value: 0.3199428097039019
|
| 21 |
+
---
|
| 22 |
+
|
| 23 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 24 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 25 |
+
|
| 26 |
+
# wav2vec2-xls-r-164m-id
|
| 27 |
+
|
| 28 |
+
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.
|
| 29 |
+
It achieves the following results on the evaluation set:
|
| 30 |
+
- Loss: 0.3215
|
| 31 |
+
- Wer: 0.3199
|
| 32 |
+
|
| 33 |
+
## Model description
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Intended uses & limitations
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training and evaluation data
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training procedure
|
| 46 |
+
|
| 47 |
+
### Training hyperparameters
|
| 48 |
+
|
| 49 |
+
The following hyperparameters were used during training:
|
| 50 |
+
- learning_rate: 0.0002
|
| 51 |
+
- train_batch_size: 24
|
| 52 |
+
- eval_batch_size: 24
|
| 53 |
+
- seed: 42
|
| 54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
+
- lr_scheduler_type: linear
|
| 56 |
+
- lr_scheduler_warmup_ratio: 0.3
|
| 57 |
+
- num_epochs: 40.0
|
| 58 |
+
- mixed_precision_training: Native AMP
|
| 59 |
+
|
| 60 |
+
### Training results
|
| 61 |
+
|
| 62 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 63 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 64 |
+
| 3.5445 | 0.92 | 1000 | 3.0106 | 1.0000 |
|
| 65 |
+
| 2.5067 | 1.84 | 2000 | 1.6134 | 0.9905 |
|
| 66 |
+
| 1.0279 | 2.75 | 3000 | 0.7667 | 0.8217 |
|
| 67 |
+
| 0.7823 | 3.67 | 4000 | 0.6141 | 0.7224 |
|
| 68 |
+
| 0.6504 | 4.59 | 5000 | 0.5228 | 0.6503 |
|
| 69 |
+
| 0.5687 | 5.51 | 6000 | 0.4666 | 0.5963 |
|
| 70 |
+
| 0.5026 | 6.43 | 7000 | 0.4288 | 0.5612 |
|
| 71 |
+
| 0.4584 | 7.35 | 8000 | 0.4048 | 0.5267 |
|
| 72 |
+
| 0.4193 | 8.26 | 9000 | 0.4057 | 0.5218 |
|
| 73 |
+
| 0.3931 | 9.18 | 10000 | 0.3820 | 0.4813 |
|
| 74 |
+
| 0.3651 | 10.1 | 11000 | 0.3686 | 0.4709 |
|
| 75 |
+
| 0.3526 | 11.02 | 12000 | 0.3665 | 0.4655 |
|
| 76 |
+
| 0.3333 | 11.94 | 13000 | 0.3440 | 0.4485 |
|
| 77 |
+
| 0.3095 | 12.86 | 14000 | 0.3314 | 0.4331 |
|
| 78 |
+
| 0.2802 | 13.77 | 15000 | 0.3360 | 0.4157 |
|
| 79 |
+
| 0.2724 | 14.69 | 16000 | 0.3331 | 0.4107 |
|
| 80 |
+
| 0.2488 | 15.61 | 17000 | 0.3255 | 0.4037 |
|
| 81 |
+
| 0.231 | 16.53 | 18000 | 0.3089 | 0.3950 |
|
| 82 |
+
| 0.2146 | 17.45 | 19000 | 0.3398 | 0.3990 |
|
| 83 |
+
| 0.2103 | 18.37 | 20000 | 0.3080 | 0.3805 |
|
| 84 |
+
| 0.2035 | 19.28 | 21000 | 0.3158 | 0.3828 |
|
| 85 |
+
| 0.1933 | 20.2 | 22000 | 0.3118 | 0.3728 |
|
| 86 |
+
| 0.1839 | 21.12 | 23000 | 0.3076 | 0.3690 |
|
| 87 |
+
| 0.1791 | 22.04 | 24000 | 0.3041 | 0.3658 |
|
| 88 |
+
| 0.1696 | 22.96 | 25000 | 0.3092 | 0.3603 |
|
| 89 |
+
| 0.1608 | 23.88 | 26000 | 0.2936 | 0.3555 |
|
| 90 |
+
| 0.1568 | 24.79 | 27000 | 0.2936 | 0.3560 |
|
| 91 |
+
| 0.1456 | 25.71 | 28000 | 0.3257 | 0.3543 |
|
| 92 |
+
| 0.1399 | 26.63 | 29000 | 0.3100 | 0.3424 |
|
| 93 |
+
| 0.1345 | 27.55 | 30000 | 0.3172 | 0.3472 |
|
| 94 |
+
| 0.1264 | 28.47 | 31000 | 0.3276 | 0.3412 |
|
| 95 |
+
| 0.1289 | 29.38 | 32000 | 0.3104 | 0.3401 |
|
| 96 |
+
| 0.1246 | 30.3 | 33000 | 0.3204 | 0.3352 |
|
| 97 |
+
| 0.1156 | 31.22 | 34000 | 0.3013 | 0.3353 |
|
| 98 |
+
| 0.1143 | 32.14 | 35000 | 0.3102 | 0.3322 |
|
| 99 |
+
| 0.1152 | 33.06 | 36000 | 0.3240 | 0.3323 |
|
| 100 |
+
| 0.1093 | 33.98 | 37000 | 0.3105 | 0.3295 |
|
| 101 |
+
| 0.101 | 34.89 | 38000 | 0.3112 | 0.3263 |
|
| 102 |
+
| 0.1017 | 35.81 | 39000 | 0.3263 | 0.3239 |
|
| 103 |
+
| 0.0915 | 36.73 | 40000 | 0.3176 | 0.3226 |
|
| 104 |
+
| 0.0943 | 37.65 | 41000 | 0.3141 | 0.3210 |
|
| 105 |
+
| 0.0898 | 38.57 | 42000 | 0.3177 | 0.3183 |
|
| 106 |
+
| 0.0923 | 39.49 | 43000 | 0.3215 | 0.3199 |
|
| 107 |
+
|
| 108 |
+
|
| 109 |
+
### Framework versions
|
| 110 |
+
|
| 111 |
+
- Transformers 4.27.0.dev0
|
| 112 |
+
- Pytorch 1.13.1+cu117
|
| 113 |
+
- Datasets 2.9.1.dev0
|
| 114 |
+
- Tokenizers 0.13.2
|