|
--- |
|
language: |
|
- tr |
|
license: apache-2.0 |
|
tags: |
|
- automatic-speech-recognition |
|
- common_voice |
|
- generated_from_trainer |
|
- tr |
|
- robust-speech-event |
|
datasets: |
|
- common_voice |
|
model-index: |
|
- name: wav2vec2-base-turkish |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: Common Voice 6.1 |
|
type: common_voice |
|
args: tr |
|
metrics: |
|
- name: Test WER |
|
type: wer |
|
value: 9.13 |
|
- name: Test CER |
|
type: cer |
|
value: 2.94 |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: Common Voice 7 |
|
type: mozilla-foundation/common_voice_7_0 |
|
args: tr |
|
metrics: |
|
- name: Test WER |
|
type: wer |
|
value: 8.88 |
|
- name: Test CER |
|
type: cer |
|
value: 2.91 |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: Common Voice 8 |
|
type: mozilla-foundation/common_voice_8_0 |
|
args: tr |
|
metrics: |
|
- name: Test WER |
|
type: wer |
|
value: 13.14 |
|
- name: Test CER |
|
type: cer |
|
value: 3.7 |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: Robust Speech Event - Dev Data |
|
type: speech-recognition-community-v2/dev_data |
|
args: tr |
|
metrics: |
|
- name: Test WER |
|
type: wer |
|
value: 40.54 |
|
- name: Test CER |
|
type: cer |
|
value: 16.9 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# |
|
|
|
This model is a fine-tuned version of [cahya/wav2vec2-base-turkish-artificial-cv](https://huggingface.co/cahya/wav2vec2-base-turkish-artificial-cv) on the COMMON_VOICE - TR dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1608 |
|
- Wer: 0.1395 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
The following datasets were used for finetuning: |
|
- [Common Voice 6.1 TR](https://huggingface.co/datasets/common_voice) 'train', 'validation', 'other' and 'invalidated' split were used for training. |
|
|
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 7.5e-06 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 24 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 2000 |
|
- num_epochs: 5.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 1.1224 | 3.45 | 500 | 0.1641 | 0.1396 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.17.0.dev0 |
|
- Pytorch 1.10.1+cu102 |
|
- Datasets 1.18.2 |
|
- Tokenizers 0.10.3 |
|
|
