metadata

license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
  - generated_from_trainer
datasets:
  - Emo-Codec/CREMA-D_synth
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: hubert-base-ls960-tone-classification
    results:
      - task:
          name: Audio Classification
          type: audio-classification
        dataset:
          name: CREMA-D
          type: Emo-Codec/CREMA-D_synth
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8016085790884718
          - name: Precision
            type: precision
            value: 0.8014677098753149
          - name: Recall
            type: recall
            value: 0.8016085790884718
          - name: F1
            type: f1
            value: 0.7989608760238184

hubert-base-ls960-tone-classification

This model is a fine-tuned version of facebook/hubert-base-ls960 on the CREMA-D dataset. It achieves the following results on the evaluation set:

Loss: 0.7499
Accuracy: 0.8016
Precision: 0.8015
Recall: 0.8016
F1: 0.7990

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-05
train_batch_size: 16
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_ratio: 0.1
num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.4326	1.0	442	1.2934	0.5147	0.5889	0.5147	0.4878
1.0447	2.0	884	0.8590	0.7051	0.7570	0.7051	0.7125
0.775	3.0	1326	0.7668	0.7426	0.7589	0.7426	0.7404
0.6593	4.0	1768	0.8127	0.7265	0.7564	0.7265	0.7245
0.5014	5.0	2210	0.8670	0.7507	0.7631	0.7507	0.7436
0.48	6.0	2652	0.7473	0.7694	0.7739	0.7694	0.7623
0.3505	7.0	3094	0.7647	0.8016	0.8039	0.8016	0.7991
0.3223	8.0	3536	0.7499	0.8016	0.8015	0.8016	0.7990

Framework versions

Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.21.0
Tokenizers 0.19.1