update model card README.md
Browse files- .ipynb_checkpoints/merges-checkpoint.txt +0 -0
- README.md +64 -0
.ipynb_checkpoints/merges-checkpoint.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
README.md
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- th
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- whisper-event
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- mozilla-foundation/common_voice_11_0
|
| 10 |
+
- google/fleurs
|
| 11 |
+
model-index:
|
| 12 |
+
- name: Whisper Medium Thai - Parinthapat Pengpun
|
| 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 |
+
# Whisper Medium Thai - Parinthapat Pengpun
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 11.0 and the FLEURS datasets.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- eval_loss: 0.1875
|
| 24 |
+
- eval_wer: 17.5807
|
| 25 |
+
- eval_cer: 8.9942
|
| 26 |
+
- eval_runtime: 14734.8594
|
| 27 |
+
- eval_samples_per_second: 0.742
|
| 28 |
+
- eval_steps_per_second: 0.046
|
| 29 |
+
- epoch: 10.02
|
| 30 |
+
- step: 11000
|
| 31 |
+
|
| 32 |
+
## Model description
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Intended uses & limitations
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training and evaluation data
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Training procedure
|
| 45 |
+
|
| 46 |
+
### Training hyperparameters
|
| 47 |
+
|
| 48 |
+
The following hyperparameters were used during training:
|
| 49 |
+
- learning_rate: 1e-05
|
| 50 |
+
- train_batch_size: 32
|
| 51 |
+
- eval_batch_size: 16
|
| 52 |
+
- seed: 42
|
| 53 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
+
- lr_scheduler_warmup_steps: 500
|
| 56 |
+
- training_steps: 15000
|
| 57 |
+
- mixed_precision_training: Native AMP
|
| 58 |
+
|
| 59 |
+
### Framework versions
|
| 60 |
+
|
| 61 |
+
- Transformers 4.26.0.dev0
|
| 62 |
+
- Pytorch 1.13.1+cu117
|
| 63 |
+
- Datasets 2.7.1.dev0
|
| 64 |
+
- Tokenizers 0.13.2
|