README.md

metadata

tags:
  - generated_from_trainer
model-index:
  - name: fusion_gttbsc_phi-3-best
    results:
      - task:
          type: dialogue act classification
        dataset:
          name: asapp/slue-phase-2
          type: hvb
        metrics:
          - name: F1 macro E2E
            type: F1 macro
            value: 70.09
          - name: F1 macro GT
            type: F1 macro
            value: 71.86
datasets:
  - asapp/slue-phase-2
language:
  - en
metrics:
  - f1-macro

fusion_gttbsc_phi-3-best

Ground truth text with prosody encoding and ASR encoding residual cross attention fusion multi-label DAC

Model description

ASR encoder: Whisper small encoder
Prosody encoder: 2 layer transformer encoder with initial dense projection
Backbone: Phi 3 mini
Fusion: 2 residual cross attention fusion layers (F_asr x F_text and F_prosody x F_text) with dense layer on top
Pooling: Self-attention
Multi-label classification head: 2 dense layers with two dropouts 0.3 and Tanh activation inbetween

Training and evaluation data

Trained on ground truth.
Evaluated on ground truth (GT) and normalized Whisper small transcripts (E2E).

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1