Mihajlo Velimirović
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mihajlo Velimirović has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Dominik Roblek
Google research
2 shared publications Beat Gfeller
1 shared publication- Beat Gfeller
Google research
1 shared publication - Blaise Aguera Y Arcas
Google Research Google Research Google Research Google Research Google Research
1 shared publication - Christian Frank
Google research
1 shared publication - Honglin Yu
Google Platforms and Services Google Deepmind Google Research Google Research Google Research
1 shared publication
Latest Publications
- Open Access
- 1 July 2026
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 160 (1) , 368-383
Cascade of asymmetric resonators with fast-acting compression compared to other cochlear models in the Auditory Modeling Toolbox
Two versions of the cascade of asymmetric resonators with fast-acting compression cochlear model have been integrated into the Auditory Modeling Toolbox and compared with nine other models there at the level of cochlear excitation patterns, linearized transfer functions, iso-response tuning curves, and other measures. Improvements to how excitation patterns and transfer functions of nonlinear models are estimated in the experiments are also explored, to help reduce confusions between linear and nonlinear effects, especially in the case of the cascade of asymmetric resonators with fast-acting compression models, which include quadratic distortion effects not modeled in the others.
- 1 July 2026
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 160 (1) , 368-383
Open AccessCascade of asymmetric resonators with fast-acting compression compared to other cochlear models in the Auditory Modeling Toolbox
Two versions of the cascade of asymmetric resonators with fast-acting compression cochlear model have been integrated into the Auditory Modeling Toolbox and compared with nine other models there at the level of cochlear excitation patterns, linearized transfer functions, iso-response tuning curves, and other measures. Improvements to how excitation patterns and transfer functions of nonlinear models are estimated in the experiments are also explored, to help reduce confusions between linear and nonlinear effects, especially in the case of the cascade of asymmetric resonators with fast-acting compression models, which include quadratic distortion effects not modeled in the others.
- Open Access
The CARFAC v2 Cochlear Model in Matlab, NumPy, and JAX
The open-source CARFAC (Cascade of Asymmetric Resonators with Fast-Acting Compression) cochlear model is upgraded to version 2, with improvements to the Matlab implementation, and with new Python/NumPy and JAX implementations -- but C++ version changes are still pending. One change addresses the DC (direct current, or zero frequency) quadratic distortion anomaly previously reported; another reduces the neural synchrony at high frequencies; the others have little or no noticeable effect in the default configuration. A new feature allows modeling a reduction of cochlear amplifier function, as a step toward a differentiable parameterized model of hearing impairment. In addition, the integration into the Auditory Model Toolbox (AMT) has been extensively improved, as the prior integration had bugs that made it unsuitable for including CARFAC in multi-model comparisons.
Open AccessThe CARFAC v2 Cochlear Model in Matlab, NumPy, and JAX
The open-source CARFAC (Cascade of Asymmetric Resonators with Fast-Acting Compression) cochlear model is upgraded to version 2, with improvements to the Matlab implementation, and with new Python/NumPy and JAX implementations -- but C++ version changes are still pending. One change addresses the DC (direct current, or zero frequency) quadratic distortion anomaly previously reported; another reduces the neural synchrony at high frequencies; the others have little or no noticeable effect in the default configuration. A new feature allows modeling a reduction of cochlear amplifier function, as a step toward a differentiable parameterized model of hearing impairment. In addition, the integration into the Auditory Model Toolbox (AMT) has been extensively improved, as the prior integration had bugs that made it unsuitable for including CARFAC in multi-model comparisons.
- Top Cited PapersOpen Access
- 20 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 28 (23299290) , 1118-1128
SPICE: Self-Supervised Pitch Estimation
We propose a model to estimate the fundamental frequency in monophonic audio, often referred to as pitch estimation. We acknowledge the fact that obtaining ground truth annotations at the required temporal and frequency resolution is a particularly daunting task. Therefore, we propose to adopt a self-supervised learning technique, which is able to estimate pitch without any form of supervision. The key observation is that pitch shift maps to a simple translation when the audio signal is analysed through the lens of the constant-Q transform (CQT). We design a self-supervised task by feeding two shifted slices of the CQT to the same convolutional encoder, and require that the difference in the outputs is proportional to the corresponding difference in pitch. In addition, we introduce a small model head on top of the encoder, which is able to determine the confidence of the pitch estimate, so as to distinguish between voiced and unvoiced audio. Our results show that the proposed method is able to estimate pitch at a level of accuracy comparable to fully supervised models, both on clean and noisy audio samples, yet it does not require access to large labeled datasets.
- 20 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 28 (23299290) , 1118-1128
Top Cited PapersOpen AccessSPICE: Self-Supervised Pitch Estimation
We propose a model to estimate the fundamental frequency in monophonic audio, often referred to as pitch estimation. We acknowledge the fact that obtaining ground truth annotations at the required temporal and frequency resolution is a particularly daunting task. Therefore, we propose to adopt a self-supervised learning technique, which is able to estimate pitch without any form of supervision. The key observation is that pitch shift maps to a simple translation when the audio signal is analysed through the lens of the constant-Q transform (CQT). We design a self-supervised task by feeding two shifted slices of the CQT to the same convolutional encoder, and require that the difference in the outputs is proportional to the corresponding difference in pitch. In addition, we introduce a small model head on top of the encoder, which is able to determine the confidence of the pitch estimate, so as to distinguish between voiced and unvoiced audio. Our results show that the proposed method is able to estimate pitch at a level of accuracy comparable to fully supervised models, both on clean and noisy audio samples, yet it does not require access to large labeled datasets.
- Open Access
Now Playing: Continuous low-power music recognition
Existing music recognition applications require a connection to a server that performs the actual recognition. In this paper we present a low-power music recognizer that runs entirely on a mobile device and automatically recognizes music without user interaction. To reduce battery consumption, a small music detector runs continuously on the mobile device's DSP chip and wakes up the main application processor only when it is confident that music is present. Once woken, the recognizer on the application processor is provided with a few seconds of audio which is fingerprinted and compared to the stored fingerprints in the on-device fingerprint database of tens of thousands of songs. Our presented system, Now Playing, has a daily battery usage of less than 1% on average, respects user privacy by running entirely on-device and can passively recognize a wide range of music.
Open AccessNow Playing: Continuous low-power music recognition
Existing music recognition applications require a connection to a server that performs the actual recognition. In this paper we present a low-power music recognizer that runs entirely on a mobile device and automatically recognizes music without user interaction. To reduce battery consumption, a small music detector runs continuously on the mobile device's DSP chip and wakes up the main application processor only when it is confident that music is present. Once woken, the recognizer on the application processor is provided with a few seconds of audio which is fingerprinted and compared to the stored fingerprints in the on-device fingerprint database of tens of thousands of songs. Our presented system, Now Playing, has a daily battery usage of less than 1% on average, respects user privacy by running entirely on-device and can passively recognize a wide range of music.