File size: 4,280 Bytes
ffe8c09 c4e91d5 ffe8c09 c4e91d5 |
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 |
---
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: xls-r-300m-hbs-ar-unfrozen-batch16
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_17_0
type: common_voice_17_0
config: hsb
split: test
args: hsb
metrics:
- name: Wer
type: wer
value: 0.46954076850984067
---
<!-- 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/badr-nlp/xlsr-continual-finetuning/runs/tkr20gft)
# xls-r-300m-hbs-ar-unfrozen-batch16
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_17_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7763
- Wer: 0.4695
- Cer: 0.1093
## 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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 100
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
| 3.3679 | 3.2258 | 100 | 3.2752 | 1.0 | 1.0 |
| 3.0469 | 6.4516 | 200 | 2.9638 | 1.0 | 0.9902 |
| 0.5512 | 9.6774 | 300 | 0.7542 | 0.7664 | 0.1947 |
| 0.3029 | 12.9032 | 400 | 0.6819 | 0.6432 | 0.1584 |
| 0.1903 | 16.1290 | 500 | 0.7312 | 0.6361 | 0.1572 |
| 0.1464 | 19.3548 | 600 | 0.7223 | 0.5916 | 0.1456 |
| 0.1205 | 22.5806 | 700 | 0.7566 | 0.5738 | 0.1416 |
| 0.091 | 25.8065 | 800 | 0.7472 | 0.5527 | 0.1308 |
| 0.0686 | 29.0323 | 900 | 0.7029 | 0.5452 | 0.1337 |
| 0.0598 | 32.2581 | 1000 | 0.7889 | 0.5464 | 0.1309 |
| 0.0607 | 35.4839 | 1100 | 0.8012 | 0.5672 | 0.1412 |
| 0.0557 | 38.7097 | 1200 | 0.7628 | 0.5302 | 0.1333 |
| 0.0421 | 41.9355 | 1300 | 0.7861 | 0.5258 | 0.1265 |
| 0.0532 | 45.1613 | 1400 | 0.7843 | 0.5314 | 0.1272 |
| 0.0298 | 48.3871 | 1500 | 0.7888 | 0.5279 | 0.1253 |
| 0.0543 | 51.6129 | 1600 | 0.7847 | 0.5295 | 0.1290 |
| 0.0404 | 54.8387 | 1700 | 0.7314 | 0.5246 | 0.1249 |
| 0.0522 | 58.0645 | 1800 | 0.7505 | 0.5134 | 0.1222 |
| 0.0275 | 61.2903 | 1900 | 0.7588 | 0.5082 | 0.1202 |
| 0.0786 | 64.5161 | 2000 | 0.7733 | 0.4930 | 0.1171 |
| 0.0439 | 67.7419 | 2100 | 0.7953 | 0.4977 | 0.1133 |
| 0.0418 | 70.9677 | 2200 | 0.7664 | 0.4897 | 0.1126 |
| 0.0399 | 74.1935 | 2300 | 0.7599 | 0.4845 | 0.1100 |
| 0.0211 | 77.4194 | 2400 | 0.7747 | 0.4763 | 0.1115 |
| 0.0225 | 80.6452 | 2500 | 0.7607 | 0.4702 | 0.1094 |
| 0.0446 | 83.8710 | 2600 | 0.7583 | 0.4768 | 0.1103 |
| 0.0236 | 87.0968 | 2700 | 0.7824 | 0.4754 | 0.1102 |
| 0.0267 | 90.3226 | 2800 | 0.7861 | 0.4726 | 0.1110 |
| 0.0255 | 93.5484 | 2900 | 0.7928 | 0.4712 | 0.1106 |
| 0.0254 | 96.7742 | 3000 | 0.7834 | 0.4684 | 0.1102 |
| 0.0137 | 100.0 | 3100 | 0.7763 | 0.4695 | 0.1093 |
### Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
|