Commit
•
afce653
1
Parent(s):
f94c5e6
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- openslr
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: wav2vec2-telugu_150
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Automatic Speech Recognition
|
| 14 |
+
type: automatic-speech-recognition
|
| 15 |
+
dataset:
|
| 16 |
+
name: openslr
|
| 17 |
+
type: openslr
|
| 18 |
+
config: SLR66
|
| 19 |
+
split: train
|
| 20 |
+
args: SLR66
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Wer
|
| 23 |
+
type: wer
|
| 24 |
+
value: 0.2212659135736059
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 28 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 29 |
+
|
| 30 |
+
# wav2vec2-telugu_150
|
| 31 |
+
|
| 32 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the openslr dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 0.3312
|
| 35 |
+
- Wer: 0.2213
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 0.0003
|
| 55 |
+
- train_batch_size: 16
|
| 56 |
+
- eval_batch_size: 8
|
| 57 |
+
- seed: 42
|
| 58 |
+
- gradient_accumulation_steps: 2
|
| 59 |
+
- total_train_batch_size: 32
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: linear
|
| 62 |
+
- lr_scheduler_warmup_steps: 500
|
| 63 |
+
- num_epochs: 150
|
| 64 |
+
- mixed_precision_training: Native AMP
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 69 |
+
|:-------------:|:------:|:-----:|:---------------:|:------:|
|
| 70 |
+
| 6.096 | 3.84 | 400 | 0.5762 | 0.7029 |
|
| 71 |
+
| 0.427 | 7.69 | 800 | 0.3124 | 0.5148 |
|
| 72 |
+
| 0.208 | 11.54 | 1200 | 0.2994 | 0.4201 |
|
| 73 |
+
| 0.1506 | 15.38 | 1600 | 0.3106 | 0.3844 |
|
| 74 |
+
| 0.1223 | 19.23 | 2000 | 0.3080 | 0.3608 |
|
| 75 |
+
| 0.1094 | 23.08 | 2400 | 0.3206 | 0.3332 |
|
| 76 |
+
| 0.0949 | 26.92 | 2800 | 0.3085 | 0.3253 |
|
| 77 |
+
| 0.0802 | 30.77 | 3200 | 0.3076 | 0.3425 |
|
| 78 |
+
| 0.0713 | 34.61 | 3600 | 0.3280 | 0.3398 |
|
| 79 |
+
| 0.0687 | 38.46 | 4000 | 0.3042 | 0.3081 |
|
| 80 |
+
| 0.0613 | 42.31 | 4400 | 0.3227 | 0.3073 |
|
| 81 |
+
| 0.0548 | 46.15 | 4800 | 0.3152 | 0.3213 |
|
| 82 |
+
| 0.0508 | 50.0 | 5200 | 0.3259 | 0.3107 |
|
| 83 |
+
| 0.0455 | 53.84 | 5600 | 0.3046 | 0.2881 |
|
| 84 |
+
| 0.0427 | 57.69 | 6000 | 0.2779 | 0.3007 |
|
| 85 |
+
| 0.0391 | 61.54 | 6400 | 0.2996 | 0.2693 |
|
| 86 |
+
| 0.0388 | 65.38 | 6800 | 0.3016 | 0.2695 |
|
| 87 |
+
| 0.0339 | 69.23 | 7200 | 0.3225 | 0.2935 |
|
| 88 |
+
| 0.0312 | 73.08 | 7600 | 0.2907 | 0.2942 |
|
| 89 |
+
| 0.029 | 76.92 | 8000 | 0.3148 | 0.3029 |
|
| 90 |
+
| 0.0254 | 80.77 | 8400 | 0.3118 | 0.2996 |
|
| 91 |
+
| 0.0229 | 84.61 | 8800 | 0.3022 | 0.2993 |
|
| 92 |
+
| 0.0231 | 88.46 | 9200 | 0.3203 | 0.2465 |
|
| 93 |
+
| 0.019 | 92.31 | 9600 | 0.3223 | 0.2460 |
|
| 94 |
+
| 0.0173 | 96.15 | 10000 | 0.3178 | 0.2501 |
|
| 95 |
+
| 0.0168 | 100.0 | 10400 | 0.2937 | 0.2415 |
|
| 96 |
+
| 0.015 | 103.84 | 10800 | 0.3062 | 0.2415 |
|
| 97 |
+
| 0.014 | 107.69 | 11200 | 0.3104 | 0.2383 |
|
| 98 |
+
| 0.012 | 111.54 | 11600 | 0.3308 | 0.2408 |
|
| 99 |
+
| 0.0111 | 115.38 | 12000 | 0.3228 | 0.2335 |
|
| 100 |
+
| 0.01 | 119.23 | 12400 | 0.3228 | 0.2374 |
|
| 101 |
+
| 0.0096 | 123.08 | 12800 | 0.3241 | 0.2304 |
|
| 102 |
+
| 0.009 | 126.92 | 13200 | 0.3237 | 0.2295 |
|
| 103 |
+
| 0.0075 | 130.77 | 13600 | 0.3221 | 0.2261 |
|
| 104 |
+
| 0.0065 | 134.61 | 14000 | 0.3310 | 0.2277 |
|
| 105 |
+
| 0.0064 | 138.46 | 14400 | 0.3348 | 0.2266 |
|
| 106 |
+
| 0.0064 | 142.31 | 14800 | 0.3330 | 0.2229 |
|
| 107 |
+
| 0.0056 | 146.15 | 15200 | 0.3310 | 0.2229 |
|
| 108 |
+
| 0.0053 | 150.0 | 15600 | 0.3312 | 0.2213 |
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
### Framework versions
|
| 112 |
+
|
| 113 |
+
- Transformers 4.24.0
|
| 114 |
+
- Pytorch 1.13.0+cu117
|
| 115 |
+
- Datasets 2.6.1
|
| 116 |
+
- Tokenizers 0.13.2
|