A Self-Supervised Model for Language Identification Integrating Phonological Knowledge
Open Access
- 14 September 2021
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 10 (18) , 2259
Abstract
In this paper, a self-supervised learning pre-trained model is proposed and successfully applied in language identification task (LID). A Transformer encoder is employed and multi-task strategy is used to train the self-supervised model: the first task is to reconstruct the masking spans of input frames and the second task is a supervision task where the phoneme and phonological labels are used with Connectionist Temporal Classification (CTC) loss. By using this multi-task learning loss, the model is expected to capture high-level speech representation in phonological space. Meanwhile, an adaptive loss is also applied for multi-task learning to balance the weight between different tasks. After the pretraining stage, the self-supervised model is used for xvector systems. Our LID experiments are carried out on the oriental language recognition (OLR) challenge data corpus and 1 s, 3 s, Full-length test sets are selected. Experimental results show that on 1 s test set, feature extraction model approach can get best performance and in 3 s, Full-length test, the fine-tuning approach can reach the best performance. Furthermore, our results prove that the multi-task training strategy is effective and the proposed model can get the best performance.Keywords
Funding Information
- Science and Technology Innovation Foundation of Shenzhen (JCYJ20180504165826861)
- National Nature Science Foundation of China (No.11590772)
This publication has 29 references indexed in Scilit:
- Cross-lingual Automatic Speech Recognition Exploiting Articulatory Features2019
- Learning Problem-Agnostic Speech Representations from Multiple Self-Supervised Tasks2019
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition2019
- Self-attention Networks for Connectionist Temporal Classification in Speech Recognition2019
- Spoken Language Recognition using X-vectors2018
- Phonetic Temporal Neural Model for Language IdentificationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2017
- Joint Estimation of Articulatory Features and Acoustic Models for Low-Resource Languages2017
- i‐vector representation based on bottleneck features for language identificationElectronics Letters, 2013
- Detection-based accented speech recognition using articulatory features2011
- Comparison of four approaches to automatic language identification of telephone speechIEEE Transactions on Speech and Audio Processing, 1996