PeopleofSuperintelligence

Researcher

Israel D. Gebru

Also known as Israel Dejene Gebru

Israel D. Gebru is a machine-learning researcher who works on audio and speech, from audio-visual speaker tracking to neural binaural sound. A bachelor's degree in computer engineering from Addis Ababa University in 2008 and a master's degree in telecommunication engineering from the University of Trento in 2013 came before a PhD in Radu Horaud's PERCEPTION team at Inria Grenoble, where they studied from 2014 to 2018. In April 2018, they joined Facebook Reality Lab in Pittsburgh, where they coauthored Neural Synthesis of Binaural Speech From Mono Audio, which received an outstanding paper award at ICLR in 2021. AudioDec followed in 2023 and BinauralFlow in 2025, with their affiliation listed as Meta's Codec Avatars Lab, and a March 2026 paper, Ethio-ASR, covered speech recognition for five Ethiopian languages and listed them with Ethiopic.ai.

Now at
Meta
Badges

Timeline

  1. 2008–2013

    Degrees in Addis Ababa and Trento

    A B.S. in Computer Engineering from Addis Ababa University comes in 2008. After working as an engineer at ZTE from 2009 to 2011, they receive an M.S. in Telecommunication Engineering from the University of Trento in 2013.

  2. 2014–2018

    PhD at Inria Grenoble

    From 2014 to 2018, they are a PhD student in the PERCEPTION team at Inria Grenoble, advised by Radu Horaud. The thesis work covers robust clustering techniques and dynamic Bayesian networks for audio-visual multiple-person localization, tracking and speech diarization.

  3. 2016–2018

    Two TPAMI papers

    IEEE Transactions on Pattern Analysis and Machine Intelligence publishes two of their papers, EM Algorithms for Weighted-Data Clustering with Applications to Audio-Visual Scene Analysis in 2016 and Audio-Visual Speaker Diarization Based on Spatiotemporal Bayesian Fusion in 2018. The diarization paper, with Sileye Ba, Xiaofei Li and Radu Horaud, combines multiple-person visual tracking with multiple speech-source localization to match speech signals to speakers.

  4. 2017

    HSCMA best paper award

    Their paper Audio-visual Tracking by Density Approximation in a Sequential Bayesian Filtering Framework wins the best paper award at the IEEE Fifth Joint Workshop on Hands-free Speech Communication and Microphone Arrays in San Francisco in March 2017. A six-month research internship at Oculus (Facebook) Research in Pittsburgh also falls in 2017.

  5. April 2018

    Joining Facebook Reality Lab

    In April 2018, they defend their PhD and join Facebook Reality Lab in Pittsburgh as a postdoctoral research scientist. Since 2019, they are a Research Scientist with Facebook Reality Labs there.

  6. 2021

    ICLR outstanding paper award

    With Alexander Richard, Dejan Markovic and four other coauthors, they present Neural Synthesis of Binaural Speech From Mono Audio at ICLR 2021. The network takes a single-channel audio source and produces two-channel binaural sound in real time, conditioned on the listener's position and orientation relative to the source. The paper receives an outstanding paper award at the conference.

  7. 2023–2025

    AudioDec, RAF and BinauralFlow

    In 2023, they coauthor AudioDec, an open-source streaming neural audio codec that reconstructs 48 kHz speech at 12 kbps with under 6 ms latency on a GPU. Real Acoustic Fields, a CVPR 2024 paper, presents a dataset of room impulse responses paired with multi-view images and 6DoF pose tracking. BinauralFlow, an ICML 2025 paper that lists them with Meta's Codec Avatars Lab, builds a causal streaming binaural speech synthesis framework on flow matching.

  8. March 2026

    Ethio-ASR and Ethiopic.ai

    In March 2026, they coauthor Ethio-ASR, a suite of multilingual speech recognition models trained on five Ethiopian languages: Amharic, Tigrinya, Oromo, Sidaama and Wolaytta. Their affiliation on the paper is Ethiopic.ai.

Career

  1. Since 2017MetaResearch Intern, Postdoctoral Research Scientist
  2. 2026Ethiopic.ai
  3. 2009–2011ZTE CorporationEngineer

Education

  1. 2018InriaPhD
  2. 2013University of TrentoM.S. in Telecommunication Engineering
  3. 2008Addis Ababa UniversityB.S. in Computer Engineering

From the biography.

  1. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Meta. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  2. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Meta. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  3. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Meta. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  4. 22 of 133 people shaping AI with a former OpenAI role now work at AnthropicIncludes Meta. Our estimates of where OpenAI recruits came from and where 133 people with former OpenAI roles work now, including the companies and labs they founded.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Israel D. Gebru yet. Send us a free-licensed photo.

  1. Israel Dejene Gebru team.inria.fr. Checked 21 October 2026.
  2. Israel D Gebru received the ICLR’21 outstanding paper award! team.inria.fr. Checked 21 October 2026.
  3. Audio-Visual Speaker Diarization Based on Spatiotemporal Bayesian Fusion team.inria.fr. Checked 21 October 2026.
  4. Neural Synthesis of Binaural Speech From Mono Audio iclr.cc. Checked 21 October 2026.
  5. Neural Synthesis of Binaural Speech from Mono Audio | ML Anthology mlanthology.org. Checked 21 October 2026.
  6. AudioDec: An Open-source Streaming High-fidelity Neural Audio Codec arxiv.org. Checked 21 October 2026.
  7. Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark openaccess.thecvf.com. Checked 21 October 2026.
  8. BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models proceedings.mlr.press. Checked 21 October 2026.
  9. BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models (arXiv HTML) arxiv.org. Checked 21 October 2026.
  10. Ethio-ASR: Joint Multilingual Speech Recognition and Language Identification for Ethiopian Languages (arXiv HTML) arxiv.org. Checked 21 October 2026.
  11. Ethio-ASR: Joint Multilingual Speech Recognition and Language Identification for Ethiopian Languages arxiv.org. Checked 21 October 2026.