Speech representations

Adventures in pre-training and fine-tuning transformers for speech technology tasks

Nik Vaessen • Boek • paperback

• Samenvatting
  Free download at https://doi.org/10.54195/9789465151090
  
  This thesis explores self-supervised speech representation learning, specifically focusing on the the wav2vec 2.0 framework.
  
  The research demonstrates that wav2vec 2.0, originally designed for automatic speech recognition, can be successfully fine-tuned for speaker recognition, even with limited labeled data. However, attempts to create a unified multi-task model for both speech and speaker recognition revealed performance trade-offs, as these tasks require orthogonal information processing.
  
  A comprehensive analysis of pre-training batch sizes shows that downstream performance primarily depends on the total amount of data observed during self-supervision. The thesis also addresses data quality requirements for self-supervised learning, finding that clean, prepared speech is essential - particularly avoiding vocal music content which causes training divergence.
  
  Finally, the research presents the creation of a 55,000-hour Dutch speech dataset from television broadcasts, demonstrating that mono-lingual pre-training can outperform multi-lingual pre-training for Dutch speech recognition.
• Productinformatie
  Binding : Paperback
  Distributievorm : Boek (print, druk)
  Formaat : 170mm x 240mm
  Aantal pagina's : 142
  Uitgeverij : Radboud University Press
  ISBN : 9789465151090
  Datum publicatie : 09-2025
  
• Inhoudsopgave
  niet beschikbaar
• Reviews (0 uit 0 reviews)
  Wil je meer weten over hoe reviews worden verzameld? Lees onze uitleg hier.