Search papers, labs, and topics across Lattice.
This study addresses the challenge of phoneme recognition in children's speech by introducing an age-aware training approach that predicts both the phoneme sequence and the learner's age. The lightweight model, with only 94M parameters, outperformed larger models like WavLM Large (317M parameters) on the DrivenData competition dataset, achieving a competitive character error rate (CER) within 0.04 of ensemble models with significantly more parameters. The resulting application, PhonemeTrainer, is designed for edge processing on modern smartphones, enhancing Automated Speech Recognition (ASR) and pronunciation assistance for children while ensuring privacy and compliance.
A lightweight model that predicts children's age alongside phonemes can outperform larger models, revolutionizing phoneme recognition in children's speech.
Detecting phonemes from children's speech has historically been difficult due to the scarcity of training data, and unique characteristics of children's speech. During a phoneme detection competition, we found that training a lightweight model to predict the age of the learner, as well as the phoneme sequence, enabled a 94M-parameter model to outperform WavLM Large models (317M) on the target DrivenData distribution, and fall within approximately 0.04 CER of competition ensembles with 90 times the parameters. This has enabled the creation of PhonemeTrainer, an application that can run on most modern cellular phones. This will ultimately enable better Automated Speech Recognition (ASR) and pronunciation helper apps for children's speech, with the privacy and compliance benefits that come with edge processing.
