Name: Audio Information Research Lab
Type: Organization
Bio: We work on computer audition, i.e., designing computational systems that can analyze and understand sounds including music, speech, and environmental sounds.
Location: Rochester, New York
Blog: https://labsites.rochester.edu/air/index.html
Audio Information Research Lab's Projects
Implementation of the paper "One-class Learning towards Generalized Voice Spoofing Detection"
Machine learning tools and framework for automatic music transcription.
Official implementation of our ASVspoof 2021 paper, "UR Channel-Robust Synthetic Speech Detection System for ASVspoof 2021"
Official implementation of our MMSP 2022 paper, "Dynamic vision-guided speaker embedding for audio-visual speaker diarization"
The code for the TMM paper "Speech Driven Talking Face Generation from a Single Image and an Emotion Condition"
Codebase for "Transcription free filler word detection with Neural semi-CRFs" [ICASSP2023]
Open source code for the paper 'Music Source Separation with Generative Flow'
Demo page
Code for the paper "A Data-Driven Methodology for Considering Feasibility and Pairwise Likelihood in Deep Learning Based Guitar Tablature Transcription Systems".
Official implementation of the handbook chapter "Generalizing Voice Presentation Attack Detection to Unseen Synthetic Attacks and Channel Variation"
Official implementation of the ICASSP 2023 paper "HRTF Field: Unifying Measured HRTF Magnitude Representation with Neural Fields"
Official Implementation of our WASPAA 2023 paper "Mitigating Cross-Database Differences for Learning Unified HRTF Representation"
Invited talk at group meeting of AIR lab
Official Implementation of our ICASSP 2023 paper "SAMO: SPEAKER ATTRACTOR MULTI-CENTER ONE-CLASS LEARNING FOR VOICE ANTI-SPOOFING"
Official implementation of the Odyssey paper "A Probabilistic Fusion Framework for Spoofing Aware Speaker Verification"
Code for the paper "Learning Sparse Analytic Filters for Piano Transcription".
Y-vector: Multiscale Waveform Encoder for Speaker Embedding