Bambara
Collection
Collection of all models trained on the Bambara language
•
11 items
•
Updated
bambara-1-hours-bambara-asr-hf
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7420
- Wer: 0.6528
- Cer: 0.2937
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.1128 | 5.71 | 200 | 1.1737 | 0.7170 | 0.3240 |
| 0.9466 | 11.43 | 400 | 1.3777 | 0.6729 | 0.3018 |
| 0.7191 | 17.14 | 600 | 1.5054 | 0.6833 | 0.3096 |
| 0.624 | 22.86 | 800 | 1.6121 | 0.6704 | 0.3018 |
| 0.4667 | 28.57 | 1000 | 1.7420 | 0.6528 | 0.2937 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.0+cu118
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 4
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
HF Inference API was unable to determine this model's library.
Model tree for asr-africa/bambara-1-hours-bambara-asr-hf-plus-lm
Base model
facebook/wav2vec2-xls-r-300m