Recent Additions

• Multi-decoder DPRNN high accuracy source counting and separation 

  Junzhe, Zhu (2020-12)

  We propose an end-to-end trainable approach to single channel speech separation with unknown number of speakers. Our approach extends the MulCat source separation backbone with additional output heads: a count-head to ...

  application/pdf

  PDF (215kB)
• Predicting from aggregated data 

  Maaz, Jedh (2020-12)

  Aggregated data, which refers to a collection of data summarized from multiple sources, is a technique commonly used in different fields of research including healthcare, web application, and sensor network. Aggregated ...

  application/pdf

  PDF (861kB)
• Spatio-temporal reprojection for virtual and augmented reality applications 

  Finn, Sinclair (2020-12)

  Late-stage reprojection is a vital tool for reducing power and compute requirements and improving perceptual latency in extended reality (XR) systems. While rotational reprojection is a well-studied topic for which several ...

  application/pdf

  PDF (5MB)
• A general method for the creation of new electronic musical instruments and a new electronic wind instrument 

  Taki, Daniel (2020-12)

  This thesis investigates a general method for the creation of new electronic musical instruments and uses that method to create a new type of electronic wind instrument. This method involves the enumeration of possible ...

  application/pdf

  PDF (3MB)
• Nvidia fellowship proposal - Sketching: a Cognitively inspired Compositional Theorem Prover that Learns 

  Brando Miranda (2020-09-11)

  Vision & motivation: Theorem proving has important applications in hardware and software verification. In hardware verification, it has been used for integrated circuit design [7, 8]. For software verification, a major ...

  application/pdf

  PDF (111kB)