tiantiaf/whisper-large-v3-msp-podcast-emotion-dim

whisper-large v3 for Dimensional Emotion Classification

Model Description

This model includes the implementation of dimensional emotion classification described in Vox-Profile: A Speech Foundation Model Benchmark for Characterizing Diverse Speaker and Speech Traits (https://arxiv.org/pdf/2505.14648)

The training pipeline used is also the top-performing solution (SAILER) in INTERSPEECH 2025—Speech Emotion Challenge (https://lab-msp.com/MSP-Podcast_Competition/IS2025/). Note that we did not use the transcript compared to our official challenge submission system, and we created a speech-only system to make the model simple but still effective.

We use the MSP-Podcast data to train this model, noting that the model might be sensitive to content information when making emotion predictions. However, this could be a good feature for classifying emotions from online content.

The outputs are in the range of 0-1 for:

Arousal (0 calm, 1 active), 
Valence (0 negative, 1 positive), 
Dominance (0 weak, 1 strong)

• Library: https://github.com/tiantiaf0627/vox-profile-release

How to use this model

Download repo

git clone [email protected]:tiantiaf0627/vox-profile-release.git

Install the package

conda create -n vox_profile python=3.8
cd vox-profile-release
pip install -e .

Load the model

# Load libraries
import torch
import torch.nn.functional as F
from src.model.emotion.whisper_emotion_dim import WhisperWrapper
# Find device
device = torch.device("cuda") if torch.cuda.is_available() else "cpu"
# Load model from Huggingface
model = WhisperWrapper.from_pretrained("tiantiaf/whisper-large-v3-msp-podcast-emotion-dim").to(device)
model.eval()

Prediction

# Load data, here just zeros as the example
# Our training data filters output audio shorter than 3 seconds (unreliable predictions) and longer than 15 seconds (computation limitation)
# So you need to prepare your audio to a maximum of 15 seconds, 16kHz and mono channel
max_audio_length = 15 * 16000
data = torch.zeros([1, 16000]).float().to(device)[:, :max_audio_length]
arousal, valence, dominance = model(data)

If you have any questions, please contact: Tiantian Feng ([email protected])

Kindly cite our paper if you are using our model or find it useful in your work

@article{feng2025vox,
  title={Vox-Profile: A Speech Foundation Model Benchmark for Characterizing Diverse Speaker and Speech Traits},
  author={Feng, Tiantian and Lee, Jihwan and Xu, Anfeng and Lee, Yoonjeong and Lertpetchpun, Thanathai and Shi, Xuan and Wang, Helin and Thebaud, Thomas and Moro-Velazquez, Laureano and Byrd, Dani and others},
  journal={arXiv preprint arXiv:2505.14648},
  year={2025}
}