# Low Resource German ASR with Untranscribed Data Spoken by Non-native Children -- INTERSPEECH 2021 Shared Task SPAPL System

arXiv: https://arxiv.org/abs/2106.09963 · HTML: https://mlsanity.com/paper/2106.09963/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 18 Jun 2021.

## Builds on (most cited first, 18 of 18)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [An Unsupervised Autoregressive Model for Speech Representation Learning](https://mlsanity.com/paper/1904.03240/) (2019)
- [Self-Training for End-to-End Speech Recognition](https://mlsanity.com/paper/1909.09116/) (2019)
- [End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures](https://mlsanity.com/paper/1911.08460/) (2019)
- [Unsupervised Pre-training of Bidirectional Speech Encoders via Masked Reconstruction](https://mlsanity.com/paper/2001.10603/) (2020)
- [Improving Transformer-based Speech Recognition Using Unsupervised Pre-training](https://mlsanity.com/paper/1910.09932/) (2019)
- [Semi-Supervised Learning with Data Augmentation for End-to-End ASR](https://mlsanity.com/paper/2007.13876/) (2020)
- [TLT-school: a Corpus of Non Native Children Speech](https://mlsanity.com/paper/2001.08051/) (2020)
- [Fundamental Frequency Feature Normalization and Data Augmentation for Child Speech Recognition](https://mlsanity.com/paper/2102.09106/) (2021)
- [Data Augmentation For Children's Speech Recognition -- The "Ethiopian" System For The SLT 2021 Children Speech Recognition Challenge](https://mlsanity.com/paper/2011.04547/) (2020)
- [Large scale weakly and semi-supervised learning for low-resource video ASR](https://mlsanity.com/paper/2005.07850/) (2020)
- [PyKaldi2: Yet another speech toolkit based on Kaldi and PyTorch](https://mlsanity.com/paper/1907.05955/) (2019)
- [Exploiting semi-supervised training through a dropout regularization in end-to-end speech recognition](https://mlsanity.com/paper/1908.05227/) (2019)
- [Exploring the Use of an Unsupervised Autoregressive Model as a Shared Encoder for Text-Dependent Speaker Verification](https://mlsanity.com/paper/2008.03615/) (2020)
- [Bi-APC: Bidirectional Autoregressive Predictive Coding for Unsupervised Pre-training and Its Application to Children's ASR](https://mlsanity.com/paper/2102.06816/) (2021)
- [The NTNU System at the Interspeech 2020 Non-Native Children's Speech ASR Challenge](https://mlsanity.com/paper/2005.08433/) (2020)
- [Data augmentation using prosody and false starts to recognize non-native children's speech](https://mlsanity.com/paper/2008.12914/) (2020)

## Built on by (most cited first, 4 of 4)

- [Transfer Learning for Robust Low-Resource Children's Speech ASR with Transformers and Source-Filter Warping](https://mlsanity.com/paper/2206.09396/) (2022)
- [Children's Speech Recognition through Discrete Token Enhancement](https://mlsanity.com/paper/2406.13431/) (2024)
- [Speech Recognition for Automatically Assessing Afrikaans and isiXhosa Preschool Oral Narratives](https://mlsanity.com/paper/2501.06478/) (2025)
- [Towards Better Meta-Initialization with Task Augmentation for Kindergarten-aged Speech Recognition](https://mlsanity.com/paper/2202.12326/) (2022)

## Abstract (arXiv)

This paper describes the SPAPL system for the INTERSPEECH 2021 Challenge: Shared Task on Automatic Speech Recognition for Non-Native Children's Speech in German. ~ 5 hours of transcribed data and ~ 60 hours of untranscribed data are provided to develop a German ASR system for children. For the training of the transcribed data, we propose a non-speech state discriminative loss (NSDL) to mitigate the influence of long-duration non-speech segments within speech utterances. In order to explore the use of the untranscribed data, various approaches are implemented and combined together to incrementally improve the system performance. First, bidirectional autoregressive predictive coding (Bi-APC) is used to learn initial parameters for acoustic modelling using the provided untranscribed data. Second, incremental semi-supervised learning is further used to iteratively generate pseudo-transcribed data. Third, different data augmentation schemes are used at different training stages to increase the variability and size of the training data. Finally, a recurrent neural network language model (RNNLM) is used for rescoring. Our system achieves a word error rate (WER) of 39.68% on the evaluation data, an approximately 12% relative improvement over the official baseline (45.21%).