Article

Training neural networks to recognize speech increased their correspondence to the human auditory pathway but did not yield a shared hierarchy of acoustic features

2021-01-27

Abstract excerpt

The correspondence between the activity of artificial neurons in convolutional neural networks (CNNs) trained to recognize objects in images and neural activity collected throughout the primate visual system has been well documented. Shallower layers of CNNs are typically more similar to early visual areas and deeper layers tend to be more similar to later visual areas, providing evidence for a shared representati...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 296b4c17-ab08-5695-8084-39c5b528af3a
DOI: 10.1101/2021.01.26.428323

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Training neural networks to recognize speech increased their correspondence to the human auditory pathway but did not yield a shared hierarchy of acoustic featuresDOI 10.1101/2021.01.26.428323Read

Spatiotemporal Dynamics of Sound Representations reveal a Hierarchical Progression of Category SelectivitySemantic proximity Read

Neural coding of spectrotemporal modulations in the auditory cortex supports speech and music categorizationSemantic proximity Read

Neural Decoding of Inferior Colliculus Multiunit Activity for Sound Category identification with temporal correlation and deep learningSemantic proximity Read

Functional architecture of cerebral cortex during naturalistic movie-watchingSemantic proximity Read

Cortical response to naturalistic stimuli is largely predictable with deep neural networksSemantic proximity Read

Brain-like dynamics in speech representations can emerge through self-supervised learningSemantic proximity Read

Reconstructing Voice Identity from Noninvasive Auditory Cortex RecordingsSemantic proximity Read

A Neural Speech Decoding Framework Leveraging Deep Learning and Speech SynthesisSemantic proximity Read

Shared Hierarchical Representations Explain Temporal Correspondence Between Brain Activity and Deep Neural NetworksSemantic proximity Read

Separable neural representations of sound sources: Speaker identity and musical timbreSemantic proximity Read

Relating dynamic brain states to dynamic machine states: human and machine solutions to the speech recognition problemSemantic proximity Read

A computational architecture incorporating shallow brain networks: integrating parallel cortical and subcortical processingSemantic proximity Read