Ivan J. Tashev

Also known as Ivan Tashev

Ivan Tashev is an audio researcher whose work connects machine learning with speech enhancement and brain–computer interfaces. He trained at the Technical University of Sofia, taught in its Faculty of Electronic Engineering and Technology, and joined Microsoft in 1998 before moving to Microsoft Research in 2001. His work included Kinect’s audio pipeline, a 2009 book on sound capture, and neural speech enhancement that reached Teams in 2020. As of October 2026, he leads the Audio and Acoustics Research Group at Microsoft Research in Redmond and is an affiliate professor at the University of Washington.

Now at: University of Washington and Microsoft Research
Works: 196
Citations: 3,968
h-index: 28
Badges: Bridge
Loyal

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

By 1998

Sofia degrees and teaching

He grows up in Yambol, Bulgaria, with two history teachers as parents, and earns an electronics engineering diploma at the Technical University of Sofia in 1984. In the university’s Research and Development Department, he works as an R&D engineer from 1984 to 1986 and as a researcher from 1986 to 1989; he completes a computer science PhD in 1990. From 1989 to 1998, he teaches data and signal processing and real-time systems programming as an assistant professor in the Faculty of Electronic Engineering and Technology.
1998–2001

Joining Microsoft Research

In 1998, he joins Microsoft and moves to Washington State. He works in Microsoft engineering teams before moving to Microsoft Research in 2001.
2009

Sound capture textbook

In 2009, Wiley publishes his book Sound Capture and Processing: Practical Approaches. It covers microphone arrays, noise reduction, acoustic echo cancellation and methods for capturing speech for communication and speech recognition.
2010

Audio processing for Kinect

He helps develop the audio pipeline released in Kinect in 2010 and designs surround-sound echo cancellation. The system removes loudspeaker sound to provide a cleaner signal for speech recognition.
Since 2012

Affiliate professor at Washington

In 2012, he becomes an affiliate professor in the Department of Electrical Engineering at the University of Washington.
2018

Neural networks for speech enhancement

In 2018, he coauthors Convolutional-Recurrent Neural Networks for Speech Enhancement with Han Zhao, Shuayb Zarar and Chin-Hui Lee. The model combines convolutional and recurrent networks to use local patterns across frequency and time in speech signals.
2020

Speech enhancement in Teams

In 2020, his work with the Teams team brings a neural-network-based speech enhancement algorithm into the product. The collaboration takes around four years to meet product requirements for memory, processing power and reliability.
2021

IEEE Fellow

He becomes an IEEE Fellow in 2021.
2023

The Industrial Innovation Award

In 2023, he receives the IEEE Signal Processing Society’s 2022 Industrial Innovation Award for contributions to microphone arrays and speech enhancement systems.
As of October 2026

Audio research at Microsoft

As of October 2026, he is a partner software architect at Microsoft Research in Redmond, where he leads the Audio and Acoustics Research Group. His research includes speech enhancement, spatial audio and brain–computer interfaces.