Open-set speaker identification in broadcast news

In this paper, we examine the problem of text-independent open-set speaker identification (OS-SI) in broadcast news. Particularly, the impact of the population of registered speakers to OS-SI performance is investigated, which is the central issue for designing practical OS-SI system. We amend the maximum mutual information (MMI)-based discriminative training scheme to facilitate its incorporation in OS-SI systems. We also improve the implementation to allow the application of MMI-based approach with 2048-component Gaussian mixture models. All systems are evaluated using NIST RT-03, RT-04 and FBIS corpora, with a maximum of 82 registered speakers. Our study shows that notable performance improvement can be obtained with MMI-based discriminative training, which reduces the equal error rate (EER) by 15.9% relatively, in comparison to the GMM-MAP scheme.

• Publication year

  2011

• Venue

  IEEE International Conference on Acoustics, Speech, and Signal Processing

• Publication date

  2011-05-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/ICASSP.2011.5947549
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• The MITLL NIST LRE 2015 Language Recognition System

  2016cited by this paper
• Discriminative In-Set/Out-of-Set Speaker Recognition

  2007influential reference
• Discriminative Training for Large-Vocabulary Speech Recognition Using Minimum Classification Error

  2007influential reference
• A discriminative training approach for text-independent speaker recognition

  2005influential reference
• Analysis of multitarget detection for speaker and language recognition

  2004influential reference
• Speaker Verification Using Adapted Gaussian Mixture Models

  2000influential reference
• Score Normalization for Text-Independent Speaker Verification Systems

  2000cited by this paper
• Text-Independent Speaker Identification

  1999cited by this paper
• Discriminative training of GMM using a modified EM algorithm for speaker recognition

  1998influential reference
• The DET curve in assessment of detection task performance

  1997cited by this paper

• VoxWatch: An open-set speaker recognition benchmark on VoxCeleb

  2023cites this paper
• Push the Limit of Adversarial Example Attack on Speaker Recognition in Physical Domain

  2022influential citation
• On Open-Set Speaker Identification with I-Vectors

  2020cites this paper
• Experiments on Open-Set Speaker Identification with Discriminatively Trained Neural Networks

  2019cites this paper
• MCE 2018: The 1st Multi-target Speaker Detection and Identification Challenge Evaluation

  2019cites this paper
• Age and Gender Identification by Indian Multilingual Speech Sample

  2019cites this paper
• Open-set Speaker Identification

  2018cites this paper
• Investigation into the Use of WFSTs and DNNs for Speech Activity Detection in Broadcast Data Transcription

  2016cites this paper
• Automatic Speaker Age Estimation and Gender Dependent Emotion Recognition

  2015cites this paper
• A Methodology for Efficient Gender Dependent Speaker Age and Emotion Identification System

  2015cites this paper
• Methodology for Gender Identification, Classification and Recognition of Human Age

  2015cites this paper
• Speaker diarization experiments for Romanian parliamentary speech

  2015cites this paper
• Improving speaker identification in TV-shows using person name detection in overlaid text and speech

  2013cites this paper
• Development of a Speaker Recognition Solution in Vidispine

  2013cites this paper