Detail publikace

Fine-Tuning Self-Supervised Models for Language Identification Using Orthonormal Constraint

PRASAD, A. CAROFILIS, A. VANDERREYDT, G. KHALIL, D. MADIKERI, S. MOTLÍČEK, P. SCHUEPBACH, C.

Originální název

Fine-Tuning Self-Supervised Models for Language Identification Using Orthonormal Constraint

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

angličtina

Originální abstrakt

Self-supervised models trained with high linguistic diversity, such as the XLS-R model, can be effectively fine-tuned for the language recognition task. Typically, a back-end classifier followed by statistics pooling layer are added during train- ing. Commonly used back-end classifiers require a large num- ber of parameters to be trained, which is not ideal in limited data conditions. In this work, we explore smaller parame- ter back-ends using factorized Time Delay Neural Network (TDNN-F). The TDNN-F architecture is also integrated into Emphasized Channel Attention, Propagation and Aggregation- TDNN (ECAPA-TDNN) models, termed ECAPA-TDNN-F, reducing the number of parameters by 30 to 50% absolute, with competitive accuracies and no change in minimum cost. The results show that the ECAPA-TDNN-F can be extended to tasks where ECAPA-TDNN is suitable. We also test the effectiveness of a linear classifier and a variant, the Orthonor- mal linear classifier, previously used in x-vector type systems. The models are trained with NIST LRE17 data and evalu- ated on NIST LRE17, LRE22 and the ATCO2 LID datasets. Both linear classifiers outperform conventional back-ends with improvements in accuracy between 0.9% and 9.1%

Klíčová slova

Language Identification, Transformers, Wav2Vec2, fine-tuning, low-resource, out-of-domain,

Autoři

PRASAD, A.; CAROFILIS, A.; VANDERREYDT, G.; KHALIL, D.; MADIKERI, S.; MOTLÍČEK, P.; SCHUEPBACH, C.

Vydáno

14. 4. 2024

Nakladatel

IEEE Signal Processing Society

Místo

Seoul

ISBN

979-8-3503-4485-1

Kniha

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Strany od

11921

Strany do

11925

Strany počet

5

URL

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446751

BibTex

@inproceedings{BUT193354,
  author="PRASAD, A. and CAROFILIS, A. and VANDERREYDT, G. and KHALIL, D. and MADIKERI, S. and MOTLÍČEK, P. and SCHUEPBACH, C.",
  title="Fine-Tuning Self-Supervised Models for Language Identification Using Orthonormal Constraint",
  booktitle="ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
  year="2024",
  pages="11921--11925",
  publisher="IEEE Signal Processing Society",
  address="Seoul",
  doi="10.1109/ICASSP48485.2024.10446751",
  isbn="979-8-3503-4485-1",
  url="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446751"
}

Dokumenty

prasad_icassp2024_fine-tuning.pdf