Search Results - "Pelecanos, Jason"

Speaker age estimation on conversational telephone speech using senone posterior based i-vectors by Sadjadi, Seyed Omid, Ganapathy, Sriram, Pelecanos, Jason W.

“…Automatic age estimation from speech has a variety of applications including natural human-computer interaction, targeted advertising, customer-agent pairing…”
Get full text

Nearest neighbor discriminant analysis for language recognition by Sadjadi, Seyed Omid, Pelecanos, Jason W., Ganapathy, Sriram

“…Many state-of-the-art i-vector based voice biometric systems use linear discriminant analysis (LDA) as a post-processing stage to increase the computational…”
Get full text

Nearest neighbor based i-vector normalization for robust speaker recognition under unseen channel conditions by Weizhong Zhu, Sadjadi, Seyed Omid, Pelecanos, Jason W.

“…Many state-of-the-art speaker recognition engines use i-vectors to represent variable-length acoustic signals in a fixed low-dimensional total variability…”
Get full text

Unsupervised channel adaptation for language identification using co-training by Ganapathy, Sriram, Omar, Mohamed, Pelecanos, Jason

“…Language identification (LID) of speech signals in conditions like adverse radio communication channel is a challenging problem. In this paper, we address the…”
Get full text

Enhancing Frequency Shifted Speech Signals in Single Side-Band Communication by Ganapathy, Sriram, Pelecanos, Jason

“…The spectral quality of speech signals communicated over high-frequency single side band (HF-SSB) radio channels is affected by acoustic artifacts like linear…”
Get full text

A Bayesian Attention Neural Network Layer for Speaker Recognition by Zhu, Weizhong, Pelecanos, Jason

“…Neural network based attention modeling has found utility in areas such as visual analysis, speech recognition and more recently speaker recognition. Attention…”
Get full text

Online speaker diarization using adapted i-vector transforms by Weizhong Zhu, Pelecanos, Jason

“…Many speaker diarization systems operate in an off-line mode. Such systems typically find homogeneous segments and then cluster these segments according to…”
Get full text

USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models by Zhao, Guanlong, Wang, Yongqiang, Pelecanos, Jason, Zhang, Yu, Liao, Hank, Huang, Yiling, Lu, Han, Wang, Quan

“…We introduce a multilingual speaker change detection model (USM-SCD) that can simultaneously detect speaker turns and perform ASR for 96 languages. This model…”
Get full text

Using Polynomial Kernel Support Vector Machines for Speaker Verification by Yaman, S., Pelecanos, J.

“…In this letter, we propose a discriminative modeling approach for the speaker verification problem that uses polynomial kernel support vector machines…”
Get full text

Feature normalization for speaker verification in room reverberation by Ganapathy, Sriram, Pelecanos, Jason, Omar, Mohamed Kamal

“…The performance of a typical speaker verification system degrades significantly in reverberant environments. This degradation is partly due to the conventional…”
Get full text

A novel approach to detecting non-native speakers and their native language by Omar, Mohamed Kamal, Pelecanos, Jason

“…Speech contains valuable information regarding the traits of speakers. This paper investigates two aspects of this information. The first is automatic…”
Get full text

Speaker diarization: A perspective on challenges and opportunities from theory to practice by Church, Kenneth, Weizhong Zhu, Vopicka, Josef, Pelecanos, Jason, Dimitriadis, Dimitrios, Fousek, Petr

“…This paper discusses some challenges and opportunities in developing a speaker diarization system for operation on real world call center telephony data. We…”
Get full text

Unifying PLDA and polynomial kernel SVMS by Yaman, Sibel, Pelecanos, Jason, Weizhong Zhu

“…Probabilistic linear discriminant analysis (PLDA) is a generative model to explain between and within class variations. When the underlying latent variables…”
Get full text

Synth2Aug: Cross-Domain Speaker Recognition with TTS Synthesized Speech by Huang, Yiling, Chen, Yutian, Pelecanos, Jason, Wang, Quan

“…In recent years, Text-To-Speech (TTS) has been used as a data augmentation technique for speech recognition to help complement inadequacies in the training…”
Get full text

Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition by Pelecanos, Jason, Wang, Quan, Moreno, Ignacio Lopez

“…Many neural network speaker recognition systems model each speaker using a fixed-dimensional embedding vector. These embeddings are generally compared using…”
Get full text

Parameter-Free Attentive Scoring for Speaker Verification by Pelecanos, Jason, Wang, Quan, Huang, Yiling, Moreno, Ignacio Lopez

“…This paper presents a novel study of parameter-free attentive scoring for speaker verification. Parameter-free scoring provides the flexibility of comparing…”
Get full text

USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models by Zhao, Guanlong, Wang, Yongqiang, Pelecanos, Jason, Zhang, Yu, Liao, Hank, Huang, Yiling, Lu, Han, Wang, Quan

“…We introduce a multilingual speaker change detection model (USM-SCD) that can simultaneously detect speaker turns and perform ASR for 96 languages. This model…”
Get full text

SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System by Chojnacka, Roza, Pelecanos, Jason, Wang, Quan, Moreno, Ignacio Lopez

“…In this paper, we describe SpeakerStew - a hybrid system to perform speaker verification on 46 languages. Two core ideas were explored in this system: (1)…”
Get full text

Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech by Huang, Yiling, Chen, Yutian, Pelecanos, Jason, Wang, Quan

“…In recent years, Text-To-Speech (TTS) has been used as a data augmentation technique for speech recognition to help complement inadequacies in the training…”
Get full text

Keyword-conditioned phone N-gram modeling with contextual information for speaker verification by Han, K. J., Pelecanos, J., Omar, M. K.

“…In this paper we present our current work on automatic speaker recognition using keyword-conditioned phone N-gram modeling. We propose the use of contextual…”
Get full text