Commit
·
310dc81
1
Parent(s):
236f82b
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- fi
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- mozilla-foundation/common_voice_7_0
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- common_voice
|
| 11 |
+
model-index:
|
| 12 |
+
- name: wav2vec2-large-xls-r-300m-finnish
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# wav2vec2-large-xls-r-300m-finnish
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - FI dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.2307
|
| 24 |
+
- Wer: 0.2984
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 7e-05
|
| 44 |
+
- train_batch_size: 32
|
| 45 |
+
- eval_batch_size: 32
|
| 46 |
+
- seed: 42
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_steps: 500
|
| 50 |
+
- num_epochs: 70.0
|
| 51 |
+
- mixed_precision_training: Native AMP
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 57 |
+
| 2.9032 | 4.39 | 500 | 2.8768 | 1.0 |
|
| 58 |
+
| 1.5724 | 8.77 | 1000 | 0.5638 | 0.6438 |
|
| 59 |
+
| 1.1818 | 13.16 | 1500 | 0.3338 | 0.4759 |
|
| 60 |
+
| 1.0798 | 17.54 | 2000 | 0.2876 | 0.4086 |
|
| 61 |
+
| 1.0296 | 21.93 | 2500 | 0.2694 | 0.4248 |
|
| 62 |
+
| 1.0014 | 26.32 | 3000 | 0.2626 | 0.3733 |
|
| 63 |
+
| 0.9616 | 30.7 | 3500 | 0.2391 | 0.3294 |
|
| 64 |
+
| 0.9303 | 35.09 | 4000 | 0.2352 | 0.3218 |
|
| 65 |
+
| 0.9248 | 39.47 | 4500 | 0.2351 | 0.3207 |
|
| 66 |
+
| 0.8837 | 43.86 | 5000 | 0.2341 | 0.3103 |
|
| 67 |
+
| 0.8887 | 48.25 | 5500 | 0.2311 | 0.3115 |
|
| 68 |
+
| 0.8529 | 52.63 | 6000 | 0.2230 | 0.3001 |
|
| 69 |
+
| 0.8404 | 57.02 | 6500 | 0.2279 | 0.3054 |
|
| 70 |
+
| 0.8242 | 61.4 | 7000 | 0.2298 | 0.3006 |
|
| 71 |
+
| 0.8288 | 65.79 | 7500 | 0.2333 | 0.2997 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
### Framework versions
|
| 75 |
+
|
| 76 |
+
- Transformers 4.16.0.dev0
|
| 77 |
+
- Pytorch 1.10.1+cu102
|
| 78 |
+
- Datasets 1.17.1.dev0
|
| 79 |
+
- Tokenizers 0.11.0
|