logo Idiap Research Institute        
 [BibTeX] [Marc21]
Implicit phonetic information modeling for speech emotion recognition
Type of publication: Conference paper
Citation: Purohit_INTERSPEECH_2023
Publication status: Accepted
Booktitle: Interspeech
Year: 2023
Month: August
Publisher: ISCA
Location: Dublin, Ireland
Crossref: Idiap-Internal-RR-07-2023
Abstract: This study investigates the efficacy of utilizing embedding spaces to model phonetic information in emotion utterances for speech emotion recognition. Our approach involves implicit modeling of phone information by deriving phone-based embeddings from networks specifically trained for phone recognition and pre-trained models fine-tuned for phone/character recognition. The results from evaluating our approach on three speech emotion databases, using both intra-corpus and inter-corpus evaluation methods demonstrate the competitive performance of implicit modeling of phonetic information compared to knowledge-based handcrafted features.
Keywords: fine-tuning, Phonetic information, self-supervised learning, Speech Emotion Recognition
Projects EMIL
Authors Purohit, Tilak
Vlasenko, Bogdan
Magimai.-Doss, Mathew
Added by: [UNK]
Total mark: 0
Attachments
  • Purohit_INTERSPEECH_2023.pdf
Notes