Work

Self-supervised learning for colloquial Finnish

Year: 2024
Type: dissertation
Abstract: Self-supervised learning has shown great results for automatic speech recognition (ASR) systems as they utilize untranscribed speech data in their training, learning speech representations from raw audio data. In this work, the aim was to create a self-supervised speech model for colloquial Finnish and compare the results of pretraining from scratch and continued pretraining. To achieve this, three different models following the base Wav2Vec 2.0 architecture were pretrained. One model was pretrained from scratch while the other two were continuously pretrained. For the continued pretraining experiments, a monolingual Finnish model and a multilingual model were used. The experiments showed similar results for the model pretrained from scratch and the continuously pretrained monolingual Finnish model, with both yielding promising results. The model pretrained continuously from the monolingual Finnish model slightly outperformed the model pretrained from scratch, obtaining a word error rate of 28.1% and a character error rate of 7.5%.
Author: Sõrmus, Birgit
Institution: Aalto University
Language: English
Open Access status: green
OpenAlex ID: W7024733530
Cites: 39
Cited by:
Related to: 10