Saved in:
Bibliographic Details
Main Authors: Li, Jialu, Hasegawa-Johnson, Mark, Karahalios, Karrie
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2309.07287
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916276295696384
author Li, Jialu
Hasegawa-Johnson, Mark
Karahalios, Karrie
author_facet Li, Jialu
Hasegawa-Johnson, Mark
Karahalios, Karrie
contents The assessment of children at risk of autism typically involves a clinician observing, taking notes, and rating children's behaviors. A machine learning model that can label adult and child audio may largely save labor in coding children's behaviors, helping clinicians capture critical events and better communicate with parents. In this study, we leverage Wav2Vec 2.0 (W2V2), pre-trained on 4300-hour of home audio of children under 5 years old, to build a unified system for tasks of clinician-child speaker diarization and vocalization classification (VC). To enhance children's VC, we build a W2V2 phoneme recognition system for children under 4 years old, and we incorporate its phonetically-tuned embeddings as auxiliary features or recognize pseudo phonetic transcripts as an auxiliary task. We test our method on two corpora (Rapid-ABC and BabbleCor) and obtain consistent improvements. Additionally, we outperform the state-of-the-art performance on the reproducible subset of BabbleCor. Code available at https://huggingface.co/lijialudew
format Preprint
id arxiv_https___arxiv_org_abs_2309_07287
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Enhancing Child Vocalization Classification with Phonetically-Tuned Embeddings for Assisting Autism Diagnosis
Li, Jialu
Hasegawa-Johnson, Mark
Karahalios, Karrie
Audio and Speech Processing
Sound
The assessment of children at risk of autism typically involves a clinician observing, taking notes, and rating children's behaviors. A machine learning model that can label adult and child audio may largely save labor in coding children's behaviors, helping clinicians capture critical events and better communicate with parents. In this study, we leverage Wav2Vec 2.0 (W2V2), pre-trained on 4300-hour of home audio of children under 5 years old, to build a unified system for tasks of clinician-child speaker diarization and vocalization classification (VC). To enhance children's VC, we build a W2V2 phoneme recognition system for children under 4 years old, and we incorporate its phonetically-tuned embeddings as auxiliary features or recognize pseudo phonetic transcripts as an auxiliary task. We test our method on two corpora (Rapid-ABC and BabbleCor) and obtain consistent improvements. Additionally, we outperform the state-of-the-art performance on the reproducible subset of BabbleCor. Code available at https://huggingface.co/lijialudew
title Enhancing Child Vocalization Classification with Phonetically-Tuned Embeddings for Assisting Autism Diagnosis
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2309.07287