Abner Hernandez

About Me

• Reasearcher for the Pattern Recognition Lab at Friedrich-Alexander-Universität Erlangen-Nürnberg and previous member of the Spoken Langauge Processing Lab.
• My research interests include in speech recoginition, particularly, impaired speech recoginition, along with impaired speech detection. However, I'm also interested in natural language processing and a variety of machine and deep learning applications in the field of linguistics and speech science.

Education

MA Linguistics, Seoul National Univiersity 2020

- Focus on computational linguistics and phonetics.

BA (Honours) Linguistics, Simon Fraser University 2017

- Focus on phonetics and neurolinguistics.

Small Projects

These projects were mostly done during my masters program at Seoul National University

UASpeech ASR kaldi baseline

This is a baseline kaldi script for a GMM-HMM based acoustic model using the UA-Speech database which is a database for dysarthric speech research. I have also included some options for data augmentation. (note that my DNN-based model using pytorch-kaldi is based on alignments from this model)

Pytorch-Kaldi with the UAspeech database

This is an example of how to create a configuration file using the pytorch-kaldi toolkit to improve dysarthric speech recognition.

Voice cloning for Dysarthric Speech

This is some experimenting I did with voice cloning using two methods PSOLA, and transfer learning TTS. Audio Samples

Dysarthric speech classification with Fricatives

This is the code used for my 2019 conference paper "Dysarthria Classification Using Acoustic Properties of Fricatives".

Publications

• Hernandez, A.; Pérez, P.A, Nöth, E., Orozco, J.R.; Maier, A.; Yang, S.H. (2022) Cross-lingual Self-Supervised Speech Representations for Improved Dysarthric Speech Recognition. To appear in Proc. Interspeech 2022. PDF LINK
• Hernandez, A.; Yang S.H. (2021) Multimodal Corpus Analysis of Autoblog 2020: Lecture Videos in Machine Learning. In: Karpov A., Potapova R. (eds) Speech and Computer. SPECOM 2021. Lecture Notes in Computer Science, vol 12997. PDF LINK
• Hernandez, A.; Kim, S.; Chung, M. Prosody-Based Measures for Automatic Severity Assessment of Dysarthric Speech. Applied Sciences 2020, 10, 6999. PDF LINK
• Hernandez, A., Yeo, E.J., Kim, S., Chung, M. (2020) Dysarthria Detection and Severity Assessment Using Rhythm-Based Metrics. Proc. Interspeech 2020, 2897-2901. PDF LINK
• Hernandez, A.; Chung, M.H. Dysarthria Classification Using Acoustic Properties of Fricatives. In Proceedings of the Seoul International Conference on Speech Sciences (SICSS), Seoul, Korea, 15–16 November 2019; pp. 43–44. PDF LINK
• Hernandez, A.; Lee, H.Y.; Chung, M.H. Acoustic analysis of fricatives in dysarthric speakers with cerebral palsy. Phonetics and Speech Sciences 2019, 11, 23–29. PDF LINK


• Sindel, A.; Hernandez, A; Yang, S.H; Christlein, V.; Maier, A. SliTraNet: Automatic Detection of Slide Transitions in Lecture Videos using Convolutional Neural Networks. Proceedings of the OAGM Workshop 2021. Computer Vision and Pattern Analysis Across Domains, 59–64. PDF LINK