Edge Phoneme Recognition for Children's Speech through Age-Aware Training

• URL: http://arxiv.org/abs/2608.10206v1
• Date: Mon, 10 Aug 2026 20:20:56 GMT
• Title: Edge Phoneme Recognition for Children's Speech through Age-Aware Training
• Abstract summary: We train a lightweight model to predict the age of the learner, as well as the phoneme sequence.<n>This has enabled the creation of PhonemeTrainer, an application that can run on most modern cellular phones.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Benchmarking Training Paradigms, Dataset Composition, and Model Scaling for Child ASR in ESPnet [72.53502346791814]
  We compare flat-start training across datasets, SSL representations (WavLM, XEUS), and decoder architectures.<n> SSL representations are biased toward adult speech, with flat-start training on child speech mitigating these biases.<n>Age-related ASR and speaker verification analysis highlights the limitations of proprietary models.
  arXiv  Detail & Related papers  (2025-08-22T17:59:35Z)
• Self-Supervised Models for Phoneme Recognition: Applications in Children's Speech for Reading Learning [9.670752318129326]
  We first compare wav2vec 2.0, HuBERT and WavLM models adapted to phoneme recognition in French child speech.<n>We then adapt it by unfreezing its transformer blocks during fine-tuning on child speech.<n>We show that WavLM base+ is more robust to various reading tasks and noise levels.
  arXiv  Detail & Related papers  (2025-03-06T18:57:16Z)
• Distilling an End-to-End Voice Assistant Without Instruction Training Data [53.524071162124464]
  Distilled Voice Assistant (DiVA) generalizes to Question Answering, Classification, and Translation. We show that DiVA better meets user preferences, achieving a 72% win rate compared with state-of-the-art models like Qwen 2 Audio.
  arXiv  Detail & Related papers  (2024-10-03T17:04:48Z)
• Child Speech Recognition in Human-Robot Interaction: Problem Solved? [0.024739484546803334]
  We revisit a study on child speech recognition from 2017 and show that indeed performance has increased. Newcomer OpenAI Whisper doing markedly better than leading commercial cloud services. While transcription is not perfect yet, the best model recognises 60.3% of sentences correctly barring small grammatical differences.
  arXiv  Detail & Related papers  (2024-04-26T13:14:28Z)
• End-to-End real time tracking of children's reading with pointer network [11.362681035467121]
  We show how a real time reading tracker can be built efficiently for children's voices. We employ a pointer network that directly learns to predict positions in the ground truth text conditioned on the streaming speech. Our best model can accurately track adult speech with 87.8% accuracy and the much harder and disfluent children's speech with 77.1% accuracy on CMU Kids data.
  arXiv  Detail & Related papers  (2023-10-17T16:12:18Z)
• Weakly-supervised Automated Audio Captioning via text only training [1.504795651143257]
  We propose a weakly-supervised approach to train an AAC model assuming only text data and a pre-trained CLAP model. We evaluate our proposed method on Clotho and AudioCaps datasets demonstrating its ability to achieve a relative performance of up to $83%$ compared to fully supervised approaches.
  arXiv  Detail & Related papers  (2023-09-21T16:40:46Z)
• Voxtlm: unified decoder-only models for consolidating speech recognition/synthesis and speech/text continuation tasks [61.3055230762097]
  We propose a decoder-only language model, VoxtLM, that can perform four tasks: speech recognition, speech synthesis, text generation, and speech continuation. VoxtLM integrates text vocabulary with discrete speech tokens from self-supervised speech features and uses special tokens to enable multitask learning.
  arXiv  Detail & Related papers  (2023-09-14T03:13:18Z)
• SpeechLM: Enhanced Speech Pre-Training with Unpaired Textual Data [100.46303484627045]
  We propose a cross-modal Speech and Language Model (SpeechLM) to align speech and text pre-training with a pre-defined unified representation. Specifically, we introduce two alternative discrete tokenizers to bridge the speech and text modalities. We evaluate SpeechLM on various spoken language processing tasks including speech recognition, speech translation, and universal representation evaluation framework SUPERB.
  arXiv  Detail & Related papers  (2022-09-30T09:12:10Z)
• Few-Shot Cross-Lingual TTS Using Transferable Phoneme Embedding [55.989376102986654]
  This paper studies a transferable phoneme embedding framework that aims to deal with the cross-lingual text-to-speech problem under the few-shot setting. We propose a framework that consists of a phoneme-based TTS model and a codebook module to project phonemes from different languages into a learned latent space.
  arXiv  Detail & Related papers  (2022-06-27T11:24:40Z)
• Mixed-Phoneme BERT: Improving BERT with Mixed Phoneme and Sup-Phoneme Representations for Text to Speech [104.65639892109381]
  We propose MixedPhoneme BERT, a novel variant of the BERT model that uses mixed phoneme and sup-phoneme representations to enhance the learning capability. Experiment results demonstrate that our proposed Mixed-Phoneme BERT significantly improves the TTS performance with 0.30 CMOS gain compared with the FastSpeech 2 baseline.
  arXiv  Detail & Related papers  (2022-03-31T17:12:26Z)
• Real-time low-resource phoneme recognition on edge devices [0.0]
  This paper shows how to create and train models for speech recognition in any language. It allows training models to recognize any language and deploying them on edge devices such as mobile phones or car displays for fast real-time speech recognition.
  arXiv  Detail & Related papers  (2021-03-25T17:34:59Z)
• Towards Zero-shot Learning for Automatic Phonemic Transcription [82.9910512414173]
  A more challenging problem is to build phonemic transcribers for languages with zero training data. Our model is able to recognize unseen phonemes in the target language without any training data. It achieves 7.7% better phoneme error rate on average over a standard multilingual model.
  arXiv  Detail & Related papers  (2020-02-26T20:38:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.