UMUTeam/w2v-bert-beto-mean-emotion-en is an English multimodal emotion recognition model developed as part of speech-emotion, an open-source multilingual and multimodal toolkit for emotion recognition from speech, text, and multimodal inputs.
This model performs multimodal emotion classification from English speech and text inputs.
The model combines acoustic representations extracted with Wav2Vec2-BERT and linguistic representations generated with RoBERTa using a mean fusion multimodal strategy.
It is designed to jointly exploit complementary emotional information from speech and text in order to improve emotion recognition performance compared to unimodal approaches.
The model predicts one of the following emotion labels:
angry
disgust
fear
happy
neutral
sad
surprise
Intended use
This model is intended for research and applied scenarios involving multimodal emotion recognition in English, such as:
multimodal conversational analysis
speech and text emotion analysis
affective computing research
emotion-aware conversational systems
human-computer interaction
multimodal AI research
The model is particularly useful in scenarios where both speech audio and transcribed text are available.
It can be used through the speech-emotion toolkit.
Out-of-scope use
This model should not be used as the sole basis for high-stakes decisions, including but not limited to:
clinical diagnosis
mental health assessment
employment, legal, or educational decisions
biometric profiling or surveillance
automated decisions affecting individuals without human oversight
Emotion recognition is inherently uncertain and context-dependent. Predictions should be interpreted as model estimates, not as definitive assessments of a person's emotional state.
Training data
The model was trained on the English multimodal datasets used in the speech-emotion project.
The training data combines multiple publicly available English speech and multimodal emotion recognition datasets, including:
RAVDESS
TESS
MELD
datasets derived from prior speech emotion recognition research benchmarks
Because the original datasets use different emotion taxonomies, all datasets were harmonized into a unified seven-class emotion taxonomy:
angry
disgust
fear
happy
neutral
sad
surprise
For the English multimodal emotion recognition setup, the same aligned speech-text samples were used for both the acoustic and textual modalities:
Training samples: 3,622
Validation samples: 453
Test samples: 453
More details about the dataset preprocessing and label harmonization pipeline are available in the project repository:
The model was evaluated on the English held-out test set used in the speech-emotion toolkit.
Performance comparison on English emotion recognition
Configuration
Accuracy
Weighted Precision
Weighted F1
Macro F1
Speech-only
95.1435
95.2700
95.1575
95.1679
Text-only
76.0842
75.5723
75.6852
68.0266
Multimodal (Concat)
96.0462
96.0880
96.0257
96.0462
Multimodal (Mean)
90.2870
90.5162
90.2334
90.2589
Multimodal (Multihead)
93.1567
93.2715
93.1898
93.2115
The results show that combining acoustic and linguistic representations improves emotion recognition performance compared to unimodal speech-only or text-only systems.
The mean fusion strategy provides competitive multimodal performance while maintaining a simpler fusion mechanism compared to more complex attention-based architectures.
How to use
Install the toolkit:
pip install speech-emotion
Multimodal emotion recognition using audio and text
python
1from speech_emotion import predict_emotion
23emotion = predict_emotion(4 audio_path="audio.wav",5 text="I was really happy to see you again.",6 language="en",7 mode="mean",8 model_config_path="model.json"9)1011print("Detected emotion:", emotion)
Multimodal emotion recognition using automatic transcription (Whisper)
If no transcription is provided, the toolkit can automatically generate it using Whisper before performing emotion recognition.
The model is designed for English multimodal emotion recognition and may not generalize reliably to other languages.
It predicts a single label from a fixed set of seven emotions.
Emotion expression is subjective and highly context-dependent.
Performance may decrease with noisy audio, inaccurate transcriptions, overlapping speakers, or domain shifts.
The model assumes that audio and text inputs are semantically aligned.
Errors in automatic speech transcription may negatively affect multimodal performance.
Bias and ethical considerations
Emotion recognition systems may reflect biases present in their training data, including differences related to accents, speaking styles, demographics, recording conditions, or annotation subjectivity.
Users should avoid interpreting predictions as objective truths about a person's internal emotional state. The model should be used with transparency, appropriate consent, and human oversight, especially in sensitive contexts.
Citation
If you use this model in your research, please cite the following works:
speech-emotion toolkit
bibtex
1@article{PAN2026102677,
2title = {speech-emotion: A multilingual and multimodal toolkit for emotion recognition from speech},
3journal = {SoftwareX},
4volume = {34},
5pages = {102677},
6year = {2026},
7issn = {2352-7110},
8doi = {https://doi.org/10.1016/j.softx.2026.102677},
9url = {https://www.sciencedirect.com/science/article/pii/S235271102600169X},
10author = {Ronghao Pan and Tomás Bernal-Beltrán and José Antonio García-Díaz and Rafael Valencia-García},
11}
Acknowledgments
This work is part of the research project LaTe4PoliticES (PID2022-138099OB-I00), funded by MICIU/AEI/10.13039/501100011033 and the European Regional Development Fund (ERDF/EU - FEDER/UE), “A way of making Europe”.
Mr. Tomás Bernal-Beltrán is supported by the University of Murcia through the predoctoral programme.
