End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
license: apache-2.0
|
| 6 |
+
base_model: openai/whisper-large-v3
|
| 7 |
+
tags:
|
| 8 |
+
- wft
|
| 9 |
+
- whisper
|
| 10 |
+
- automatic-speech-recognition
|
| 11 |
+
- audio
|
| 12 |
+
- speech
|
| 13 |
+
- generated_from_trainer
|
| 14 |
+
datasets:
|
| 15 |
+
- ntnu-smil/sandi2025-ds
|
| 16 |
+
model-index:
|
| 17 |
+
- name: whisper-large-v3-sandi-train-dev-1
|
| 18 |
+
results: []
|
| 19 |
+
---
|
| 20 |
+
|
| 21 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 22 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 23 |
+
|
| 24 |
+
# whisper-large-v3-sandi-train-dev-1
|
| 25 |
+
|
| 26 |
+
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the ntnu-smil/sandi2025-ds dataset.
|
| 27 |
+
It achieves the following results on the evaluation set:
|
| 28 |
+
- eval_loss: 1.3669
|
| 29 |
+
- eval_wer: 70.9647
|
| 30 |
+
- eval_cer: 206.0000
|
| 31 |
+
- eval_decode_runtime: 293.2787
|
| 32 |
+
- eval_wer_runtime: 0.2383
|
| 33 |
+
- eval_cer_runtime: 0.5705
|
| 34 |
+
- eval_runtime: 777.5153
|
| 35 |
+
- eval_samples_per_second: 4.179
|
| 36 |
+
- eval_steps_per_second: 0.131
|
| 37 |
+
- epoch: 1.0357
|
| 38 |
+
- step: 7
|
| 39 |
+
|
| 40 |
+
## Model description
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Intended uses & limitations
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training and evaluation data
|
| 49 |
+
|
| 50 |
+
More information needed
|
| 51 |
+
|
| 52 |
+
## Training procedure
|
| 53 |
+
|
| 54 |
+
### Training hyperparameters
|
| 55 |
+
|
| 56 |
+
The following hyperparameters were used during training:
|
| 57 |
+
- learning_rate: 7e-05
|
| 58 |
+
- train_batch_size: 32
|
| 59 |
+
- eval_batch_size: 32
|
| 60 |
+
- seed: 42
|
| 61 |
+
- gradient_accumulation_steps: 32
|
| 62 |
+
- total_train_batch_size: 1024
|
| 63 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.98) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
|
| 64 |
+
- lr_scheduler_type: linear
|
| 65 |
+
- training_steps: 28
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- PEFT 0.15.1
|
| 70 |
+
- Transformers 4.50.3
|
| 71 |
+
- Pytorch 2.1.0+cu118
|
| 72 |
+
- Datasets 3.5.0
|
| 73 |
+
- Tokenizers 0.21.1
|