Speaker identification and verification using Gaussian mixture speaker models

By Douglas A. Reynolds. 1995

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion? (2019, background): “Indeed, state-of-the-art speech processing algorithms can infer not only the spoken contents from the speech signal, but also the speaker’s identity [5], intention [6–9], gender [10,11], emotional state [12–14], pathological condition [15–17], personality [18,19] and cultural [20,21] attributes to…”
Frame-by-frame language identification in short utterances using deep neural networks (2015, methodology): “Based on the MAP adaptation approach in a GMM framework [24], utterances in language or speaker recognition are typically represented by the accumulated zero- and centered first-order Baum-Welch statistics, N and F, respectively, computed from a Universal Background Model (UBM) λ.”
Feature warping for robust speaker verification (2001, methodology): “Thus, for clean speech using the same microphone for recordings, many of these enhancement techniques may actually reduce system performance [8].”
Unsupervised feature learning for audio classification using convolutional deep belief networks (2009, methodology): “Details: In [16], MFCC features (with multiple frames) were computed for each utterance; then a Gaussian mixture model was trained for each speaker (treating each individual MFCC frame as a input example to the GMM.”
Biometric Identification through Hand Geometry Measurements (2000, methodology): “Based on the works carried out by Reynolds and rose ([9], [10]), the probability density of a sample belonging to a class u is: px 3 =u M iI c i P L”
Speaker recognition by means of Deep Belief Networks (2013, background): “An ivector is a compac t representation of a GMM supervector [15], represent ing both the speaker and channel characteristics of a g iven speech segment, which captures most of the supervec to variability.”
An overview of text-independent speaker recognition: From features to supervectors (2010, methodology): “In the recognition phase, background speakers are used in the normalization of the speaker match score [71, 101, 139, 193, 197, 206].”
A Tutorial on Text-Independent Speaker Verification (2004, background): “have been the subject of much research [9, 10, 11, 12].”
A Multimodal Platform for Database Recording and Elderly People Monitoring (2008, methodology): “The method used by this module is based on Gaussian Mixture Model (GMM) (D.A.Reynolds, 1995) (Kmeans followed by Expectation Maximisation in 20 steps).”
Learning Speaker-Specific Characteristics With a Deep Neural Architecture (2011, background): “Thus, exploring a task-specific representation characterizing the proper information for a given task, e.g., a speaker-specific representation for SR, has posed a challenge to speech information processing for a long time [1], [3]–[7].”
Speaker recognition with recurrent neural networks (2000, methodology): “…applied to speaker recognition, including hidden Markov models [Siohan, 1998], vector quantisation [Qin et al., 1998], gaussian mixture modelling [Reynolds, 1995], multi-layer perceptrons [Altosaar and Meister, 1995], Radial Basis Functions [Finan et al., 1996] and genetic algorithms [Hannah et…”
End-to-End attention based text-dependent speaker verification (2016, unknown): “Previously, the traditional techniques [7, 8] used for textindependent speaker verification have been found ineffective for the text-dependent tasks [4].”
A Gaussian mixture model based classification scheme for myoelectric control of powered upper limb prostheses (2005, background): “The GMM has become the dominant approach in speaker recognition and verification over the past several years [16]–[19].”
Speaker identification and clustering using convolutional neural networks (2016, methodology): “However, other works achieve nearly perfect accuracy by using GMMs [28][29].”