Robust text-independent speaker identification using Gaussian mixture speaker models

By Douglas A. Reynolds, Richard C. Rose. 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

Exploring the Encoding Layer and Loss Function in End-to-End Speaker and Language Recognition System (2018, methodology): “), includes vector quantization (VQ) codebooks learned by K-means [12], a universal background model (UBM) learned by Gaussian Mixture Model (GMM) GMM [13, 14] or a supervised phonetically-aware acoustic model learned by DNN [10, 15].”
VoxCeleb: A Large-Scale Speaker Identification Dataset (2017, background): “For a long time, speaker identiﬁcation was the domain of Gaussian Mixture Models (GMMs) trained on low dimensional feature vectors [7, 8].”
A review of biometric technology along with trends and prospects (2014, background): “Table 1: Summary of hand region modalities: Modality Feature Set Recognition Techniques Datasets Fingerprint Ridge flow, ridge pattern, singular points, ridge skeleton, ridge flow, ridge ending, ridge contours, ridge kernel, orientation field, island, spur, crossover, learned feature, sweat pores, d”
Spoofing and countermeasures for speaker verification: A survey (2015, unknown): “In the classical approach (Reynolds and Rose, 1995), features X are typically MFCCs and the acoustic models are Gaussian mixture models (GMMs) (see below).”
DeepEar: robust smartphone audio sensing in unconstrained acoustic environments using deep learning (2015, background): “32-component GMMs provide classification, and are trained with a variance limiting technique [66] that lowers the impact of noisy data.”
Computer vision based method for real-time fire and flame detection (2006, unknown): “The cloud is represented by using a mixture of Gaussians in the RGB color space as described in (Reynolds and Rose, 1995).”
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”
A Gaussian mixture model based classification scheme for myoelectric control of powered upper limb prostheses (2005, background): “The GMM has been shown to generally produce the best performance in speaker identification and verification systems [17], [18].”
Speaker identification and verification using Gaussian mixture speaker models (1995, methodology): “A more detailed discussion of how GMMs apply to speaker modeling can be found in (Reynolds, 1992; Reynolds and Rose, 1995).”
The ASVspoof 2017 Challenge: Assessing the Limits of Replay Spoofing Attack Detection (2017, background): “Automatic speaker verification (ASV) [1, 2, 3] technology is used in a growing range of applications which require not only robustness to changes in the acoustic environment, but also resilience to intentional circumvention, known as spoofing [4] or, according to ISO/IEC 30107-1:2016(1) standard, pr”
A probabilistic approach for damage identification and crack mode classification in reinforced concrete structures (2013, methodology): “For a D-dimensional measurement, training, or feature vector ~x, the mixture density is defined as (Reynolds and Rose, 1995)”
Survey on speech emotion recognition: Features, classification schemes, and databases (2011, background): “At the same time, there exist speaker identification techniques with high recognition performance such as the GMM-based text-independent speaker identification system proposed by Reynolds [107].”
Voxceleb: Large-scale speaker verification in the wild (2020, unknown): “For a long time, speaker identification was dominated by Gaussian Mixture Models (GMMs) trained on low dimensional feature vectors (Reynolds et al., 2000; Reynolds and Rose, 1995).”
An Overview of Speaker Identification: Accuracy and Robustness Issues (2011, methodology): “SECOND QUARTER 2011 IEEE CIRCUITS AND SYSTEMS MAGAZINE 31 in particular [ 24 ].”