Spoken Language Identification with Pre-trained Models and Margin Loss

TL;DR

This paper introduces a language identification approach using pre-trained ECAPA-TDNN models and margin loss to improve discriminability, achieving significant accuracy gains on the TidyLang Challenge 2026 dataset.

Contribution

It presents a novel combination of pre-trained models and margin-based losses for speaker-controlled language ID, outperforming baseline methods.

Findings

Achieved 85.95% macro accuracy and 90.96% micro accuracy.

Reduced EER to 17.08%, halving the error rate.

Significantly outperformed the official baseline.

Abstract

For the speaker-controlled spoken language identification task proposed in the TidyLang Challenge 2026, this paper proposes a language identification method based on pre-trained models and margin-based losses. The proposed method adopts a pre-trained ECAPA-TDNN as the feature encoder and incorporates margin-based losses to enhance the discriminative ability of language representations, thereby improving inter-class separability and reducing the interference of non-linguistic factors such as speaker characteristics. Experimental results on the Tidy-X dataset show that the proposed method achieves 85.95% macro accuracy and 90.96% micro accuracy on the language identification task and 17.08% equal error rate (EER) on the verification task. Compared with the official baseline, the macro accuracy improves by 45.7%, the micro accuracy improves by 15.2%, and the EER is reduced by approximately 50.8%, demonstrating the effectiveness of the proposed method. The code will be released at https://github.com/PunkMale/TidyLang2026.