File size: 7,813 Bytes
d7e87bf fe8db32 d7e87bf fe8db32 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 |
---
license: apache-2.0
base_model: facebook/wav2vec2-large
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-sw-cv-20hr-v1
results: []
---
<!-- 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. -->
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/asr-africa-research-team/ASR%20Africa/runs/w0c4nymx)
# wav2vec2-large-sw-cv-20hr-v1
This model is a fine-tuned version of [facebook/wav2vec2-large](https://huggingface.co/facebook/wav2vec2-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: inf
- Model Preparation Time: 0.0059
- Wer: 0.3464
- Cer: 0.1302
## 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: 0.0001
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 100
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer | Cer |
|:-------------:|:-------:|:-----:|:---------------:|:----------------------:|:------:|:------:|
| 4.0167 | 0.9976 | 210 | 1.3039 | 0.0059 | 0.9301 | 0.3195 |
| 0.7784 | 2.0 | 421 | 0.7306 | 0.0059 | 0.5818 | 0.1788 |
| 0.5359 | 2.9976 | 631 | 0.5886 | 0.0059 | 0.5048 | 0.1517 |
| 0.427 | 4.0 | 842 | 0.5274 | 0.0059 | 0.4460 | 0.1345 |
| 0.3657 | 4.9976 | 1052 | 0.5617 | 0.0059 | 0.4620 | 0.1430 |
| 0.3219 | 6.0 | 1263 | 0.5162 | 0.0059 | 0.408 | 0.1240 |
| 0.2922 | 6.9976 | 1473 | 0.4861 | 0.0059 | 0.4074 | 0.1256 |
| 0.2681 | 8.0 | 1684 | 0.5076 | 0.0059 | 0.404 | 0.1253 |
| 0.2459 | 8.9976 | 1894 | 0.5042 | 0.0059 | 0.3915 | 0.1205 |
| 0.2332 | 10.0 | 2105 | 0.5051 | 0.0059 | 0.3706 | 0.1120 |
| 0.2181 | 10.9976 | 2315 | 0.5370 | 0.0059 | 0.3750 | 0.1149 |
| 0.2073 | 12.0 | 2526 | 0.5231 | 0.0059 | 0.3860 | 0.1249 |
| 0.1982 | 12.9976 | 2736 | 0.5290 | 0.0059 | 0.4045 | 0.1239 |
| 0.1875 | 14.0 | 2947 | 0.5184 | 0.0059 | 0.3755 | 0.1153 |
| 0.1782 | 14.9976 | 3157 | 0.5215 | 0.0059 | 0.3587 | 0.1100 |
| 0.1684 | 16.0 | 3368 | 0.5395 | 0.0059 | 0.371 | 0.1142 |
| 0.1629 | 16.9976 | 3578 | 0.5499 | 0.0059 | 0.3608 | 0.1101 |
| 0.1563 | 18.0 | 3789 | 0.5478 | 0.0059 | 0.3577 | 0.1107 |
| 0.1516 | 18.9976 | 3999 | 0.5290 | 0.0059 | 0.3649 | 0.1148 |
| 0.1431 | 20.0 | 4210 | 0.5765 | 0.0059 | 0.3657 | 0.1167 |
| 0.1366 | 20.9976 | 4420 | 0.5604 | 0.0059 | 0.3617 | 0.1137 |
| 0.1345 | 22.0 | 4631 | 0.5546 | 0.0059 | 0.3604 | 0.1118 |
| 0.1303 | 22.9976 | 4841 | 0.5284 | 0.0059 | 0.3511 | 0.1089 |
| 0.122 | 24.0 | 5052 | 0.5668 | 0.0059 | 0.3555 | 0.1111 |
| 0.1183 | 24.9976 | 5262 | 0.5874 | 0.0059 | 0.3521 | 0.1088 |
| 0.1151 | 26.0 | 5473 | 0.5539 | 0.0059 | 0.3379 | 0.1044 |
| 0.1108 | 26.9976 | 5683 | 0.6110 | 0.0059 | 0.3375 | 0.1051 |
| 0.1089 | 28.0 | 5894 | 0.5582 | 0.0059 | 0.3397 | 0.1029 |
| 0.1064 | 28.9976 | 6104 | 0.5774 | 0.0059 | 0.3432 | 0.1062 |
| 0.1026 | 30.0 | 6315 | 0.6042 | 0.0059 | 0.3420 | 0.1062 |
| 0.0983 | 30.9976 | 6525 | 0.5793 | 0.0059 | 0.3402 | 0.1046 |
| 0.0952 | 32.0 | 6736 | 0.6083 | 0.0059 | 0.3423 | 0.1074 |
| 0.0927 | 32.9976 | 6946 | 0.6015 | 0.0059 | 0.3363 | 0.1035 |
| 0.0895 | 34.0 | 7157 | 0.5790 | 0.0059 | 0.3368 | 0.1041 |
| 0.0889 | 34.9976 | 7367 | 0.5530 | 0.0059 | 0.3338 | 0.1023 |
| 0.0865 | 36.0 | 7578 | 0.5598 | 0.0059 | 0.3267 | 0.1009 |
| 0.0828 | 36.9976 | 7788 | 0.5699 | 0.0059 | 0.3249 | 0.1001 |
| 0.0814 | 38.0 | 7999 | 0.5756 | 0.0059 | 0.3237 | 0.0996 |
| 0.0819 | 38.9976 | 8209 | 0.5878 | 0.0059 | 0.3363 | 0.1052 |
| 0.077 | 40.0 | 8420 | 0.5852 | 0.0059 | 0.3216 | 0.0984 |
| 0.075 | 40.9976 | 8630 | 0.5940 | 0.0059 | 0.3295 | 0.1022 |
| 0.0725 | 42.0 | 8841 | 0.5779 | 0.0059 | 0.3219 | 0.0997 |
| 0.0701 | 42.9976 | 9051 | 0.5962 | 0.0059 | 0.3144 | 0.0965 |
| 0.0693 | 44.0 | 9262 | 0.6192 | 0.0059 | 0.317 | 0.0975 |
| 0.0659 | 44.9976 | 9472 | 0.5989 | 0.0059 | 0.3126 | 0.0964 |
| 0.0662 | 46.0 | 9683 | 0.6069 | 0.0059 | 0.3112 | 0.0975 |
| 0.0646 | 46.9976 | 9893 | 0.6309 | 0.0059 | 0.3164 | 0.0986 |
| 0.0626 | 48.0 | 10104 | 0.6266 | 0.0059 | 0.3199 | 0.1007 |
| 0.062 | 48.9976 | 10314 | 0.6403 | 0.0059 | 0.3116 | 0.0963 |
| 0.0591 | 50.0 | 10525 | 0.6140 | 0.0059 | 0.3133 | 0.0965 |
| 0.0568 | 50.9976 | 10735 | 0.5947 | 0.0059 | 0.3078 | 0.0950 |
| 0.0538 | 52.0 | 10946 | 0.6202 | 0.0059 | 0.3029 | 0.0939 |
| 0.0544 | 52.9976 | 11156 | 0.6215 | 0.0059 | 0.312 | 0.0966 |
| 0.0526 | 54.0 | 11367 | 0.6637 | 0.0059 | 0.3093 | 0.0959 |
| 0.05 | 54.9976 | 11577 | 0.6513 | 0.0059 | 0.3079 | 0.0955 |
| 0.0518 | 56.0 | 11788 | 0.6611 | 0.0059 | 0.3070 | 0.0948 |
| 0.0493 | 56.9976 | 11998 | 0.6415 | 0.0059 | 0.3041 | 0.0941 |
| 0.0482 | 58.0 | 12209 | 0.6386 | 0.0059 | 0.3042 | 0.0939 |
| 0.0461 | 58.9976 | 12419 | 0.6664 | 0.0059 | 0.316 | 0.0995 |
| 0.0445 | 60.0 | 12630 | 0.6472 | 0.0059 | 0.3057 | 0.0963 |
| 0.0449 | 60.9976 | 12840 | 0.6510 | 0.0059 | 0.3103 | 0.0972 |
| 0.0437 | 62.0 | 13051 | 0.6696 | 0.0059 | 0.3166 | 0.1005 |
### Framework versions
- Transformers 4.43.1
- Pytorch 2.2.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
|