Locally normalized filter banks applied to deep neural-network-based robust speech recognition

J Fredes, J Novoa, S King, RM Stern… - IEEE Signal …, 2017 - ieeexplore.ieee.org
This letter describes modifications to locally normalized filter banks (LNFB), which
substantially improve their performance on the Aurora-4 robust speech recognition task …

[PDF][PDF] Robustness over time-varying channels in DNN-hmm ASR based human-robot interaction.

J Novoa, J Wuth, JP Escudero, J Fredes, R Mahu… - …, 2017 - isca-archive.org
This paper addresses the problem of time-varying channels in speech-recognition-based
human-robot interaction using Locally-Normalized Filter-Bank features (LNFB), and training …

Reconocimiento robusto de voz en sistemas DNN-HMM utilizando parametrización normalizada localmente

JA Fredes Sandoval - 2023 - repositorio.uchile.cl
En esta tesis se describe las modificaciones a la parametrización de bancos de filtros
normalizados localmente que mejoran sustancialmente su rendimiento en la tarea de …

[PDF][PDF] The Use of Locally Normalized Cepstral Coefficients (LNCC) to Improve Speaker Recognition Accuracy in Highly Reverberant Rooms.

V Poblete, JP Escudero, J Fredes, J Novoa… - …, 2016 - researchgate.net
We describe the ability of LNCC features (Locally Normalized Cepstral Coefficients) to
improve speaker recognition accuracy in highly reverberant environments. We used a …

Exploring the robustness of features and enhancement on speech recognition systems in highly-reverberant real environments

J Novoa, JP Escudero, J Wuth, V Poblete… - arXiv preprint arXiv …, 2018 - arxiv.org
This paper evaluates the robustness of a DNN-HMM-based speech recognition system in
highly-reverberant real environments using the HRRE database. The performance of locally …

Robust speech recognition in noisy and reverberant environments using deep neural network-based systems

JE Novoa Ilic - 2018 - repositorio.uchile.cl
In this thesis an uncertainty weighting scheme for deep neural network-hidden Markov
model (DNN-HMM) based automatic speech recognition (ASR) is proposed to increase …