# Transcribing Children's Speech: ASR Performance and Obtaining Reliable Orthographic Transcriptions

arXiv: https://arxiv.org/abs/2605.28833 · HTML: https://mlsanity.com/paper/2605.28833/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 42 references that are in the map.
- First posted on arXiv 10 Apr 2026.

## Builds on (most cited first, 21 of 21)

- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Recent Advances in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/2111.01690/) (2021)
- [Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST](https://mlsanity.com/paper/2509.14128/) (2025)
- [Confidence Estimation for Attention-based Sequence-to-sequence Models for Speech Recognition](https://mlsanity.com/paper/2010.11428/) (2020)
- [Adaptation of Whisper models to child speech recognition](https://mlsanity.com/paper/2307.13008/) (2023)
- [An evaluation of word-level confidence estimation for end-to-end automatic speech recognition](https://mlsanity.com/paper/2101.05525/) (2021)
- [Improving child speech recognition with augmented child-like speech](https://mlsanity.com/paper/2406.10284/) (2024)
- [Evaluating ASR Confidence Scores for Automated Error Detection in User-Assisted Correction Interfaces](https://mlsanity.com/paper/2503.15124/) (2025)
- [Automatic Assessment of Oral Reading Accuracy for Reading Diagnostics](https://mlsanity.com/paper/2306.03444/) (2023)
- [Exploring Pathological Speech Quality Assessment with ASR-Powered Wav2Vec2 in Data-Scarce Context](https://mlsanity.com/paper/2403.20184/) (2024)
- [Improving the Inclusivity of Dutch Speech Recognition by Fine-tuning Whisper on the JASMIN-CGN Corpus](https://mlsanity.com/paper/2502.17284/) (2025)
- [Word-Level ASR Quality Estimation for Efficient Corpus Sampling and Post-Editing through Analyzing Attentions of a Reference-Free Metric](https://mlsanity.com/paper/2401.11268/) (2024)
- [Automatic Quality Estimation for ASR System Combination](https://mlsanity.com/paper/1706.07238/) (2017)
- [Improving Child Speech Recognition and Reading Mistake Detection by Using Prompts](https://mlsanity.com/paper/2506.11079/) (2025)
- [Utterance-level neural confidence measure for end-to-end children speech recognition](https://mlsanity.com/paper/2109.07750/) (2021)
- [An ASR-Based Tutor for Learning to Read: How to Optimize Feedback to First Graders](https://mlsanity.com/paper/2306.04190/) (2023)
- [Evaluating ASR robustness to spontaneous speech errors: A study of WhisperX using a Speech Error Database](https://mlsanity.com/paper/2508.13060/) (2025)
- [Reading Miscue Detection in Primary School through Automatic Speech Recognition](https://mlsanity.com/paper/2406.07060/) (2024)
- [Evaluation of Speech Foundation Models for ASR on Child-Adult Conversations in Autism Diagnostic Sessions](https://mlsanity.com/paper/2409.16135/) (2024)

## Abstract (arXiv)

Automatic speech recognition (ASR) has the potential to substantially reduce manual annotation effort in child speech research by generating automatic transcriptions. However, obtaining reliably high-quality ASR transcriptions for child speech remains challenging in low-resource languages due to limited child-specific pre-trained models and highly diverse noise conditions. This study investigates the effectiveness of state-of-the-art ASR models on child speech through two research questions, by evaluating nine ASR models from three model families (Whisper, Parakeet, and Wav2Vec2) on two Dutch child speech datasets, JASMIN and DART. Research question 1 examines the performance of ASR-models applied to child speech. The fine-tuned Whisper-medium model achieves the best overall performance, with a WER of 5.54% on JASMIN and 70.37% on DART, showing that the noisy DART data are clearly more challenging. Research question 2 examines to what extent it is possible to select a subset for which reliable orthographic transcriptions can be obtained automatically, without the need for manual verification. We use an utterance-level selection method that compares ASR output with the original read prompt to identify correctly pronounced recordings. Using the proposed selection method, 42.0% [for JASMIN] and 18.1% [for DART] of the utterances can be automatically identified as correctly pronounced with high confidence, resulting in very low error rates on an utterance level (precisions of 98.3% and higher) and reducing the need for manual verification.