PeopleofSuperintelligence

Researcher

Dejan Markovic

Dejan Markovic is a research scientist whose work connects neural speech with spatial audio. After completing doctoral studies in Italy, he moved to the United States in 2017 and worked as a research scientist at Facebook in Pennsylvania by July 2018. His 2021 binaural-speech paper, which turns single-channel audio into binaural sound conditioned on the listener's position and orientation, won an ICLR Outstanding Paper Award. He coauthored a CVPR 2022 audio-visual speech codec, AudioDec in 2023, and ScoreDec in 2024; in 2025, he coauthored BinauralFlow and SoundVista. By September 2026, he is a research scientist at Meta.

Now at
Meta
Works
8
Citations
84
h-index
2

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2017

    Move to the United States

    After completing doctoral studies in Italy, he moves to the United States in 2017.

  2. 2021

    An ICLR outstanding paper

    In 2021, he coauthors Neural Synthesis of Binaural Speech from Mono Audio. The network turns a single-channel source into two-channel binaural sound conditioned on the listener's relative position and orientation; the paper wins an ICLR Outstanding Paper Award.

  3. 2022

    Audio-visual speech codecs

    In 2022, he coauthors the CVPR paper Audio-Visual Speech Codecs: Rethinking Audio-Visual Speech Enhancement by Re-Synthesis. The model uses audio-visual cues to generate neural speech-codec codes and synthesize clean speech from noisy signals for AR/VR telecommunications.

  4. 2023

    The AudioDec codec

    In 2023, he coauthors AudioDec, an open-source, real-time neural audio codec that reconstructs 48 kHz speech at 12 kbps with latency under 6 ms on GPU and 10 ms on CPU.

  5. 2024

    ScoreDec adds a diffusion filter

    In 2024, he coauthors ScoreDec, which combines AudioDec with a score-based diffusion post-filter in the complex spectral domain. It encodes and decodes full-band 48 kHz speech at 24 kbps while preserving phase information.

  6. 2025

    BinauralFlow at ICML

    At ICML 2025, he coauthors BinauralFlow, a flow-matching framework for streaming binaural speech synthesis. Its causal U-Net estimates the current audio frame from past information.

  7. 2025

    SoundVista at CVPR

    At CVPR 2025, he coauthors SoundVista, a method that synthesizes ambient sound at unseen viewpoints from recordings made by sparsely distributed microphones.

  8. September 2026

    Research scientist at Meta

    By July 2018, he works as a research scientist at Facebook in Pennsylvania. In September 2026, he is a research scientist at Meta, a role he begins in July 2018.

  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 with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Dejan Markovic yet. Send us a free-licensed photo.

  1. POSJETA BRACE MARKOVIC GK U NJUJORKU gov.me. Checked 22 October 2026.
  2. Alexander Richard - Research Scientist - Meta Reality Labs Research alexanderrichard.github.io. Checked 22 October 2026.
  3. CVPR 2022 Open Access Repository openaccess.thecvf.com. Checked 22 October 2026.
  4. [2305.16608] AudioDec: An Open-source Streaming High-fidelity Neural Audio Codec arxiv.org. Checked 22 October 2026.
  5. AudioDec: An Open-source Streaming High-fidelity Neural Audio Codec bigpon.github.io. Checked 22 October 2026.
  6. [2401.12160] ScoreDec: A Phase-preserving High-Fidelity Audio Codec with A Generalized Score-based Diffusion Post-filter arxiv.org. Checked 22 October 2026.
  7. BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models proceedings.mlr.press. Checked 22 October 2026.
  8. CVPR 2025 Open Access Repository openaccess.thecvf.com. Checked 22 October 2026.
  9. Find Dejan Markovic's email & phone number - Research Scientist at Meta, Pittsburgh | SignalHire signalhire.com. Checked 22 October 2026.
  10. Reviewers | EUSIPCO 2026 eurasip.org. Checked 22 October 2026.