---
language:
- lv
license: apache-2.0
base_model: openai/whisper-medium
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_1
metrics:
- wer
model-index:
- name: Whisper medium LV - Arturs Logins
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: Common Voice 16.1
      type: mozilla-foundation/common_voice_16_1
      config: lv
      split: test
      args: 'config: lv, split: test'
    metrics:
    - name: Wer
      type: wer
      value: 15.611516648123525
---

<!-- 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 medium LV - Arturs Logins

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 16.1 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2088
- Wer: 15.6115

## 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: 5e-06
- train_batch_size: 4
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| No log        | 0.06  | 200  | 1.1684          | 30.5962 |
| 2.1727        | 0.12  | 400  | 0.3890          | 28.8540 |
| 0.6371        | 0.19  | 600  | 0.3618          | 27.1689 |
| 0.5874        | 0.25  | 800  | 0.3285          | 25.4756 |
| 0.5186        | 0.31  | 1000 | 0.3187          | 24.5665 |
| 0.5186        | 0.37  | 1200 | 0.3037          | 23.3372 |
| 0.4772        | 0.44  | 1400 | 0.2848          | 21.4594 |
| 0.4575        | 0.5   | 1600 | 0.2782          | 21.2966 |
| 0.4541        | 0.56  | 1800 | 0.2674          | 20.6670 |
| 0.4098        | 0.62  | 2000 | 0.2599          | 19.8366 |
| 0.4098        | 0.69  | 2200 | 0.2490          | 19.0877 |
| 0.4028        | 0.75  | 2400 | 0.2436          | 18.6589 |
| 0.404         | 0.81  | 2600 | 0.2383          | 18.4310 |
| 0.3723        | 0.87  | 2800 | 0.2314          | 17.7254 |
| 0.3578        | 0.94  | 3000 | 0.2299          | 17.9832 |
| 0.3578        | 1.0   | 3200 | 0.2208          | 17.3265 |
| 0.331         | 1.06  | 3400 | 0.2258          | 17.1583 |
| 0.1796        | 1.12  | 3600 | 0.2229          | 16.9548 |
| 0.1835        | 1.19  | 3800 | 0.2201          | 16.8245 |
| 0.1684        | 1.25  | 4000 | 0.2172          | 16.4934 |
| 0.1684        | 1.31  | 4200 | 0.2134          | 16.0457 |
| 0.1701        | 1.37  | 4400 | 0.2109          | 15.8557 |
| 0.172         | 1.44  | 4600 | 0.2099          | 15.7391 |
| 0.1709        | 1.5   | 4800 | 0.2094          | 15.6169 |
| 0.1704        | 1.56  | 5000 | 0.2088          | 15.6115 |


### Framework versions

- Transformers 4.40.0.dev0
- Pytorch 2.0.1
- Datasets 2.18.0
- Tokenizers 0.15.2