Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: facebook/wav2vec2-xls-r-300m
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- common_voice_17_0
|
| 9 |
+
metrics:
|
| 10 |
+
- wer
|
| 11 |
+
model-index:
|
| 12 |
+
- name: result_data-5
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Automatic Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: common_voice_17_0
|
| 19 |
+
type: common_voice_17_0
|
| 20 |
+
config: uk
|
| 21 |
+
split: test
|
| 22 |
+
args: uk
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Wer
|
| 25 |
+
type: wer
|
| 26 |
+
value: 0.6674214548542315
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 30 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 31 |
+
|
| 32 |
+
# result_data-5
|
| 33 |
+
|
| 34 |
+
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.
|
| 35 |
+
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.4794
|
| 37 |
+
- Wer: 0.6674
|
| 38 |
+
- Cer: 0.2557
|
| 39 |
+
|
| 40 |
+
## Model description
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Intended uses & limitations
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training and evaluation data
|
| 49 |
+
|
| 50 |
+
More information needed
|
| 51 |
+
|
| 52 |
+
## Training procedure
|
| 53 |
+
|
| 54 |
+
### Training hyperparameters
|
| 55 |
+
|
| 56 |
+
The following hyperparameters were used during training:
|
| 57 |
+
- learning_rate: 8.442713223799316e-05
|
| 58 |
+
- train_batch_size: 16
|
| 59 |
+
- eval_batch_size: 16
|
| 60 |
+
- seed: 42
|
| 61 |
+
- distributed_type: multi-GPU
|
| 62 |
+
- num_devices: 2
|
| 63 |
+
- total_train_batch_size: 32
|
| 64 |
+
- total_eval_batch_size: 32
|
| 65 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 66 |
+
- lr_scheduler_type: linear
|
| 67 |
+
- lr_scheduler_warmup_steps: 84
|
| 68 |
+
- num_epochs: 7.0
|
| 69 |
+
- mixed_precision_training: Native AMP
|
| 70 |
+
|
| 71 |
+
### Training results
|
| 72 |
+
|
| 73 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 74 |
+
|:-------------:|:------:|:----:|:---------------:|:------:|:------:|
|
| 75 |
+
| 3.1606 | 0.9099 | 1000 | 3.1650 | 1.0 | 0.9920 |
|
| 76 |
+
| 1.2343 | 1.8198 | 2000 | 1.0356 | 0.9474 | 0.3937 |
|
| 77 |
+
| 0.7338 | 2.7298 | 3000 | 0.6668 | 0.7973 | 0.3038 |
|
| 78 |
+
| 0.6334 | 3.6397 | 4000 | 0.5813 | 0.7560 | 0.2852 |
|
| 79 |
+
| 0.5414 | 4.5496 | 5000 | 0.5283 | 0.6952 | 0.2675 |
|
| 80 |
+
| 0.5056 | 5.4595 | 6000 | 0.5042 | 0.6821 | 0.2633 |
|
| 81 |
+
| 0.4778 | 6.3694 | 7000 | 0.4794 | 0.6674 | 0.2557 |
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
### Framework versions
|
| 85 |
+
|
| 86 |
+
- Transformers 4.49.0
|
| 87 |
+
- Pytorch 2.6.0+cu124
|
| 88 |
+
- Datasets 3.3.2
|
| 89 |
+
- Tokenizers 0.21.0
|