# Cantonese Automatic Speech Recognition Using Transfer Learning from Mandarin

arXiv: https://arxiv.org/abs/1911.09271 · HTML: https://mlsanity.com/paper/1911.09271/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Nov 2019.

## Built on by (most cited first, 2 of 2)

- [Effects of Layer Freezing on Transferring a Speech Recognition System to Under-resourced Languages](https://mlsanity.com/paper/2102.04097/) (2021)
- [textscCantoNLU: A benchmark for Cantonese natural language understanding](https://mlsanity.com/paper/2510.20670/) (2025)

## Abstract (arXiv)

We propose a system to develop a basic automatic speech recognizer(ASR) for Cantonese, a low-resource language, through transfer learning of Mandarin, a high-resource language. We take a time-delayed neural network trained on Mandarin, and perform weight transfer of several layers to a newly initialized model for Cantonese. We experiment with the number of layers transferred, their learning rates, and pretraining i-vectors. Key findings are that this approach allows for quicker training time with less data. We find that for every epoch, log-probability is smaller for transfer learning models compared to a Cantonese-only model. The transfer learning models show slight improvement in CER.