Visual Interpretation of DNN-based Acoustic Models using Deep Autoencoders

Tamás Grósz*, Mikko Kurimo

*Tämän työn vastaava kirjoittaja

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionScientificvertaisarvioitu

27 Lataukset (Pure)

Abstrakti

In the past few years, Deep Neural Networks (DNN) have become the state-of-the-art solution in several areas, including automatic speech recognition (ASR), unfortunately, they are generally viewed as black boxes. Recently, this started to change as researchers have dedicated much effort into interpreting their behavior. In this work, we concentrate on visual interpretation by depicting the hidden activation vectors of the DNN, and propose the usage of deep Autoencoders (DAE) to transform these hidden representations for inspection. We use multiple metrics to compare our approach with other, widely-used algorithms and the results show that our approach is quite competitive. The main advantage of using Autoencoders over the existing ones is that after the training phase, it applies a fixed transformation that can be used to visualize any hidden activation vector without any further optimization, which is not true for the other methods.
AlkuperäiskieliEnglanti
OtsikkoMachine Learning Methods in Visualisation for Big Data
AlaotsikkoEurographics proceedings
ToimittajatDaniel Archambault, Ian Nabney, Jaakko Peltonen
KustantajaEuropean Association for Computer Graphics
Sivut25-29
ISBN (elektroninen)978-3-03868-113-7
DOI - pysyväislinkit
TilaJulkaistu - 2020
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisuussa
TapahtumaInternational Workshop on Machine Learning in Visualisation for Big Data - Norrköping, Ruotsi
Kesto: 25 toukokuuta 202025 toukokuuta 2020

Conference

ConferenceInternational Workshop on Machine Learning in Visualisation for Big Data
LyhennettäMLVis
Maa/AlueRuotsi
KaupunkiNorrköping
Ajanjakso25/05/202025/05/2020

Sormenjälki

Sukella tutkimusaiheisiin 'Visual Interpretation of DNN-based Acoustic Models using Deep Autoencoders'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä