Keywords:
- breathing patterns
- Convolution Neural Network
- cross-transfer knowledge
- Customer satisfaction
- Emotion Recognition
- end-to-end modelling
- Expressive Vocalizations
- fine-tuning
- Formant transitions
- Foundation Model
- LoRA
- modalities fusion
- Multi-task learning
- Parkinson’s disease
- Peft
- phoneme modeling
- Phonetic information
- pre-trained embedding
- Self-supervised embedding
- self-supervised learning
- Speech Analysis
- Speech Emotion Recognition
- Speech for health
- Spoken Language Understanding
- Stop-consonants
Publications of Tilak Purohit
2024
| Cross-transfer Knowledge between Speech and Text Encoders to Evaluate Customer Satisfaction, , , , and , in: Interspeech, Kos Island, Greece, ISCA, 2024 |
|
2023
| A Study on the Importance of Formant Transitions for Stop-Consonant Classification in VCV Sequence, , , and , in: Interspeech, Dublin, Ireland, ISCA, 2023 |
|
| Implicit phonetic information modeling for speech emotion recognition, , and , in: Interspeech, Dublin, Ireland, ISCA, 2023 |
|
| Towards learning emotion information from short segments of speech, , , , and , in: International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Rhodes Island, Greece, IEEE, 2023 |
|
2022
| Comparing Biosignal and Acoustic feature Representation for Continuous Emotion Recognition, , , , and , in: International Multimodal Sentiment Analysis Workshop and Challenge, 2022 |
|
| Comparing supervised and self-supervised embedding for ExVo Multi-Task learning track, , , and , in: Proceedings of the ICML Expressive Vocalizations Workshop held in conjunction with the 39th International Conference on Machine Learning, Maryland, USA, 2022 |
|