Alert button
Picture for Jason Pelecanos

Jason Pelecanos

Alert button

USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models

Sep 14, 2023
Guanlong Zhao, Yongqiang Wang, Jason Pelecanos, Yu Zhang, Hank Liao, Yiling Huang, Han Lu, Quan Wang

Figure 1 for USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models
Figure 2 for USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models
Figure 3 for USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models
Figure 4 for USM-SCD: Multilingual Speaker Change Detection Based on Large Pretrained Foundation Models
Viaarxiv icon

Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech

Mar 21, 2022
Quan Wang, Yang Yu, Jason Pelecanos, Yiling Huang, Ignacio Lopez Moreno

Figure 1 for Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech
Figure 2 for Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech
Figure 3 for Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech
Figure 4 for Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech
Viaarxiv icon

Parameter-Free Attentive Scoring for Speaker Verification

Mar 10, 2022
Jason Pelecanos, Quan Wang, Yiling Huang, Ignacio Lopez Moreno

Figure 1 for Parameter-Free Attentive Scoring for Speaker Verification
Figure 2 for Parameter-Free Attentive Scoring for Speaker Verification
Figure 3 for Parameter-Free Attentive Scoring for Speaker Verification
Figure 4 for Parameter-Free Attentive Scoring for Speaker Verification
Viaarxiv icon

SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System

Apr 26, 2021
Roza Chojnacka, Jason Pelecanos, Quan Wang, Ignacio Lopez Moreno

Figure 1 for SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 2 for SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 3 for SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 4 for SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Viaarxiv icon

Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System

Apr 05, 2021
Roza Chojnacka, Jason Pelecanos, Quan Wang, Ignacio Lopez Moreno

Figure 1 for Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 2 for Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 3 for Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Figure 4 for Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
Viaarxiv icon

Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition

Apr 05, 2021
Jason Pelecanos, Quan Wang, Ignacio Lopez Moreno

Figure 1 for Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition
Figure 2 for Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition
Figure 3 for Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition
Figure 4 for Dr-Vectors: Decision Residual Networks and an Improved Loss for Speaker Recognition
Viaarxiv icon

Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech

Nov 24, 2020
Yiling Huang, Yutian Chen, Jason Pelecanos, Quan Wang

Figure 1 for Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech
Figure 2 for Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech
Figure 3 for Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech
Figure 4 for Synth2Aug: Cross-domain speaker recognition with TTS synthesized speech
Viaarxiv icon

VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition

Sep 09, 2020
Quan Wang, Ignacio Lopez Moreno, Mert Saglam, Kevin Wilson, Alan Chiao, Renjie Liu, Yanzhang He, Wei Li, Jason Pelecanos, Marily Nika, Alexander Gruenstein

Figure 1 for VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition
Figure 2 for VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition
Figure 3 for VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition
Viaarxiv icon

The IBM Speaker Recognition System: Recent Advances and Error Analysis

May 05, 2016
Seyed Omid Sadjadi, Jason Pelecanos, Sriram Ganapathy

Figure 1 for The IBM Speaker Recognition System: Recent Advances and Error Analysis
Figure 2 for The IBM Speaker Recognition System: Recent Advances and Error Analysis
Figure 3 for The IBM Speaker Recognition System: Recent Advances and Error Analysis
Figure 4 for The IBM Speaker Recognition System: Recent Advances and Error Analysis
Viaarxiv icon