Benchmarking Training Paradigms, Dataset Composition, and Model Scaling for Child ASR in ESPnet
This work addresses the problem of limited annotated data and acoustic variability in child speech recognition for researchers and developers, though it is incremental as it builds on existing methods.
The paper tackles child speech recognition by comparing flat-start training with fine-tuning approaches, finding that SSL representations are biased toward adult speech and flat-start training on child speech mitigates these biases, with model scaling showing consistent improvements up to 1B parameters before plateauing.
Despite advancements in ASR, child speech recognition remains challenging due to acoustic variability and limited annotated data. While fine-tuning adult ASR models on child speech is common, comparisons with flat-start training remain underexplored. We compare flat-start training across multiple datasets, SSL representations (WavLM, XEUS), and decoder architectures. Our results show that SSL representations are biased toward adult speech, with flat-start training on child speech mitigating these biases. We also analyze model scaling, finding consistent improvements up to 1B parameters, beyond which performance plateaus. Additionally, age-related ASR and speaker verification analysis highlights the limitations of proprietary models like Whisper, emphasizing the need for open-data models for reliable child speech research. All investigations are conducted using ESPnet, and our publicly available benchmark provides insights into training strategies for robust child speech processing.