---
language:
- jpn
license: apache-2.0
base_model: openai/whisper-small
tags:
- speaker-diarization
- speaker-segmentation
- generated_from_trainer
datasets:
- diarizers-community/callhome
model-index:
- name: speaker-segmentation-fine-tuned-callhome-jpn
  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. -->

# speaker-segmentation-fine-tuned-callhome-jpn

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the diarizers-community/callhome dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7482
- Der: 0.2201
- False Alarm: 0.0465
- Missed Detection: 0.1319
- Confusion: 0.0417

## 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: 0.001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Der    | False Alarm | Missed Detection | Confusion |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-----------:|:----------------:|:---------:|
| 0.5488        | 1.0   | 328  | 0.7565          | 0.2280 | 0.0461      | 0.1355           | 0.0465    |
| 0.475         | 2.0   | 656  | 0.7596          | 0.2220 | 0.0467      | 0.1334           | 0.0419    |
| 0.4734        | 3.0   | 984  | 0.7531          | 0.2215 | 0.0437      | 0.1364           | 0.0414    |
| 0.4535        | 4.0   | 1312 | 0.7468          | 0.2194 | 0.0462      | 0.1323           | 0.0409    |
| 0.4764        | 5.0   | 1640 | 0.7482          | 0.2201 | 0.0465      | 0.1319           | 0.0417    |


### Framework versions

- Transformers 4.41.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1