You need to agree to share your contact information to access this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
One more step before getting this model.
This model is open access and available only for non-commercial use, with an SIL International AI & NLP RAIL-M license further specifying rights and usage.
The SIL RAIL-M License specifies:
- You can't use the model to deliberately produce nor share illegal or harmful outputs or content. Particularly, you cannot use the model use with the intent or effect of harming or enabling discrimination against Indigenous People.
- SIL claims no rights on outputs you generate for non-commercial use, you are free to use them and are accountable for their use, which must not go against the provisions set in the license
- You may re-distribute the weights and use the model non-commercially including as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the SIL International AI & NLP RAIL-M to all your users (please read the license entirely and carefully). Please read the full license here: https://huggingface.co/spaces/sil-ai/model-license
By clicking on "Access repository" below, you accept that your contact information (email address and username) can be shared with the model authors as well.
If you would like to ask about commercial uses of this model, please email us.
Log in or Sign Up to review the conditions and access this model content.
wav2vec2-bloom-speech-mam
Model description
- Homepage: SIL AI
- Point of Contact: SIL AI email
- Source Data: Bloom Library
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the SIL-AI/bloom-speech - MAM (Mam) dataset. It achieves the following results on the evaluation set:
- Loss: 0.5938
- Wer: 0.3023
- Cer: 0.0787
Users should refer to the original model for tutorials on using a trained model for inference.
Intended uses & limitations
Users of this model must abide by the SIL RAIL-M License.
This model is created as a proof of concept and no guarantees are made regarding the performance of the model is specific situations.
Training and evaluation data
Training, Validation, and Test datasets were generated from the same corpus, ensuring that no duplicate files were used.
Training procedure
Standard finetuning of XLS-R was used based on the examples in the Hugging Face Transformers Github
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- 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: 250
- num_epochs: 1000.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| No log | 6.1 | 250 | 3.1703 | 1.0 | 1.0 |
| 4.9494 | 12.2 | 500 | 2.8022 | 1.0 | 1.0 |
| 4.9494 | 18.29 | 750 | 1.3280 | 0.8842 | 0.3435 |
| 1.5795 | 24.39 | 1000 | 0.6121 | 0.5177 | 0.1293 |
| 1.5795 | 30.49 | 1250 | 0.5740 | 0.4759 | 0.1181 |
| 0.3087 | 36.59 | 1500 | 0.4996 | 0.3601 | 0.0899 |
| 0.3087 | 42.68 | 1750 | 0.5313 | 0.3730 | 0.0887 |
| 0.1772 | 48.78 | 2000 | 0.5345 | 0.3473 | 0.0818 |
| 0.1772 | 54.88 | 2250 | 0.5637 | 0.3408 | 0.0824 |
| 0.1331 | 60.98 | 2500 | 0.5938 | 0.3023 | 0.0787 |
| 0.1331 | 67.07 | 2750 | 0.5622 | 0.3376 | 0.0824 |
| 0.1147 | 73.17 | 3000 | 0.5609 | 0.3923 | 0.0943 |
| 0.1147 | 79.27 | 3250 | 0.5213 | 0.3344 | 0.0812 |
Framework versions
- Transformers 4.21.0.dev0
- Pytorch 1.9.0+cu111
- Datasets 2.2.2
- Tokenizers 0.12.1
- Downloads last month
- 0
Evaluation results
- Test WER on Bloom Speech mamself-reported30.230
- Test CER on Bloom Speech mamself-reported7.870