File size: 1,479 Bytes
ba89106 8d0ec49 ba89106 8d0ec49 ba89106 8d0ec49 ba89106 8d0ec49 ba89106 8d0ec49 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
---
language:
- nyn
license: apache-2.0
base_model: openai/whisper-small
tags:
- whisper-event
- generated_from_trainer
datasets:
- tericlabs
model-index:
- name: Whisper Small Runyankore
results: []
---
<!-- 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. -->
# Whisper Small Runyankore
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Yogera data dataset.
It achieves the following results on the evaluation set:
- eval_loss: 0.9862
- eval_wer: 44.6947
- eval_runtime: 52.8005
- eval_samples_per_second: 4.261
- eval_steps_per_second: 0.549
- epoch: 42.55
- step: 5000
## 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: 1e-05
- 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: 1000
- training_steps: 10000
- mixed_precision_training: Native AMP
### Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
|