Whisper Activity Detection Using CNN-LSTM Based Attention Pooling Network Trained for a Speaker Identification Task

2020 ◽  
Author(s):  
Abinay Reddy Naini ◽  
Malla Satyapriya ◽  
Prasanta Kumar Ghosh
Keyword(s):  
Speaker Identification ◽  
Activity Detection ◽  
Identification Task

Improving SVM by Modifying Kernel Functions for Speaker Identification Task

2010 ◽  
Vol 4 (6) ◽  
pp. 100-105 ◽  
Author(s):  
Siwar Zribi Boujelbene ◽  
Dorra Ben Ayed Mezghani ◽  
Noureddine Ellouze
Keyword(s):  
Speaker Identification ◽  
Kernel Functions ◽  
Identification Task

Simple Visualizations Enhance Speaker Identification when Listening to Spatialized Voices

2005 ◽  
Vol 49 (4) ◽  
pp. 615-618
Author(s):  
Ryan Kilgore ◽  
Mark Chignell
Keyword(s):  
Response Time ◽  
Task Performance ◽  
Speaker Identification ◽  
Mental Workload ◽  
Visual Representation ◽  
Identification Accuracy ◽  
Spatial Audio ◽  
Identification Task ◽  
Voice Identification ◽  
Performance Response

Spatial audio has been demonstrated to enhance performance in a variety of listening tasks. The utility of visually reinforcing spatialized audio with depictions of voice locations in collaborative applications, however, has been questioned. In this experiment, we compared the accuracy, response time, confidence in task performance, and subjective mental workload of 18 participants in a voice-identification task under three different display conditions: 1) traditional mono audio; 2) spatial audio; 3) spatial audio with a visual representation of voice locations. Each format was investigated using four and eight unique stimuli voices. Results showed greater voice-identification accuracy for the spatial-plus-visual format than for the spatialand mono-only formats, and that visualization benefits increased with voice number. Spatialization was also found to increase confidence in task performance. Response time and mental workload remained unchanged across display conditions. These results indicate visualizations may benefit users of large, unfamiliar audio spaces.

Perceptual Evaluation of Speech Naturalness in Speakers of Varying Gender Identities

2020 ◽  
Vol 63 (7) ◽  
pp. 2054-2069
Author(s):  
Brandon Merritt ◽  
Tessa Bent
Keyword(s):  
Spontaneous Speech ◽  
Identification Accuracy ◽  
Rating Task ◽  
Gender Identification ◽  
Identification Task ◽  
Male And Female ◽  
Perceptual Evaluation ◽  
Speech Training ◽  
And Gender ◽  
Speech Naturalness

Purpose The purpose of this study was to investigate how speech naturalness relates to masculinity–femininity and gender identification (accuracy and reaction time) for cisgender male and female speakers as well as transmasculine and transfeminine speakers. Method Stimuli included spontaneous speech samples from 20 speakers who are transgender (10 transmasculine and 10 transfeminine) and 20 speakers who are cisgender (10 male and 10 female). Fifty-two listeners completed three tasks: a two-alternative forced-choice gender identification task, a speech naturalness rating task, and a masculinity/femininity rating task. Results Transfeminine and transmasculine speakers were rated as significantly less natural sounding than cisgender speakers. Speakers rated as less natural took longer to identify and were identified less accurately in the gender identification task; furthermore, they were rated as less prototypically masculine/feminine. Conclusions Perceptual speech naturalness for both transfeminine and transmasculine speakers is strongly associated with gender cues in spontaneous speech. Training to align a speaker's voice with their gender identity may concurrently improve perceptual speech naturalness. Supplemental Material https://doi.org/10.23641/asha.12543158

Combining the Post-Cue Task and the Perceptual Identification Task to Assess Parallel Activation and Mutual Facilitation of Related Primes and Targets

2018 ◽  
Vol 65 (2) ◽  
pp. 84-97 ◽  
Author(s):  
Demian Scherer ◽  
Dirk Wentura
Keyword(s):  
Semantic Priming ◽  
Stimulus Onset ◽  
The Other ◽  
Perceptual Identification ◽  
Identification Task ◽  
Priming Effects ◽  
Priming Paradigm ◽  
Parallel Activation ◽  
Prime Target

Abstract. Recent theories assume a mutual facilitation in case of semantic overlap for concepts being activated simultaneously. We provide evidence for this claim using a semantic priming paradigm. To test for mutual facilitation of related concepts, a perceptual identification task was employed, presenting prime-target pairs briefly and masked, with an SOA of 0 ms (i.e., prime and target were presented concurrently, one above the other). Participants were instructed to identify the target. In Experiment 1, a cue defining the target was presented at stimulus onset, whereas in Experiment 2 the cue was not presented before the offset of stimuli. Accordingly, in Experiment 2, a post-cue task was merged with the perceptual identification task. We obtained significant semantic priming effects in both experiments. This result is compatible with the view that two concepts can both be activated in parallel and can mutually facilitate each other if they are related.

Sign in / Sign up

Export Citation Format

Share Document