# Wavelet Scattering Transform for Improving Generalization in Low-Resourced Spoken Language Identification

arXiv: https://arxiv.org/abs/2310.00602 · HTML: https://mlsanity.com/paper/2310.00602/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 31 references that are in the map.
- First posted on arXiv 1 Oct 2023.

## Builds on (most cited first, 12 of 12)

- [ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification](https://mlsanity.com/paper/2005.07143/) (2020)
- [Additive Margin Softmax for Face Verification](https://mlsanity.com/paper/1801.05599/) (2018)
- [VoxLingua107: a Dataset for Spoken Language Recognition](https://mlsanity.com/paper/2011.12998/) (2020)
- [Deep Scattering Spectrum](https://mlsanity.com/paper/1304.6763/) (2013)
- [AP20-OLR Challenge: Three Tasks and Their Baselines](https://mlsanity.com/paper/2006.03473/) (2020)
- [Time-Frequency Scattering Accurately Models Auditory Similarities Between Instrumental Playing Techniques](https://mlsanity.com/paper/2007.10926/) (2020)
- [Data Quality as Predictor of Voice Anti-Spoofing Generalization](https://mlsanity.com/paper/2103.14602/) (2021)
- [Audio Texture Synthesis with Scattering Moments](https://mlsanity.com/paper/1311.0407/) (2013)
- [Cross-Corpora Language Recognition: A Preliminary Investigation with Indian Languages](https://mlsanity.com/paper/2105.04639/) (2021)
- [Cross-Corpora Spoken Language Identification with Domain Diversification and Generalization](https://mlsanity.com/paper/2302.05110/) (2023)
- [Deep scattering network for speech emotion recognition](https://mlsanity.com/paper/2105.04806/) (2021)
- [An Overview of Indian Spoken Language Recognition from Machine Learning Perspective](https://mlsanity.com/paper/2212.03812/) (2022)

## Abstract (arXiv)

Commonly used features in spoken language identification (LID), such as mel-spectrogram or MFCC, lose high-frequency information due to windowing. The loss further increases for longer temporal contexts. To improve generalization of the low-resourced LID systems, we investigate an alternate feature representation, wavelet scattering transform (WST), that compensates for the shortcomings. To our knowledge, WST is not explored earlier in LID tasks. We first optimize WST features for multiple South Asian LID corpora. We show that LID requires low octave resolution and frequency-scattering is not useful. Further, cross-corpora evaluations show that the optimal WST hyper-parameters depend on both train and test corpora. Hence, we develop fused ECAPA-TDNN based LID systems with different sets of WST hyper-parameters to improve generalization for unknown data. Compared to MFCC, EER is reduced upto 14.05% and 6.40% for same-corpora and blind VoxLingua107 evaluations, respectively.