Audio signal processing
Our research topics and projects are related to headset signal processing, loudspeakers, sound synthesis, effects processing algorithms, digital filters, musical instruments, acoustic measurements, subjective evaluation, and interactive audio systems. The emphasis of our research is on digital signal processing and machine learning methods which account for the properties of human hearing. We have collaborated with several companies, such as Genelec and Nokia Technologies. Much of our academic research is conducted in collaboration with foreign partners, such as Aalborg University, The University of Edinburgh, and Stanford University.
This research group organizes the Audio Signal Processing course at Aalto University annually in January-April. We also arrange the Audio Technology Seminar every second year (February-May). Both courses are suitable for Master's and doctoral students, who have a solid background in acoustics and digital signal processing. Additionally, we are responsible for the Noise Control course for Master's and doctoral students and for the Sound and Speech Processing course (Äänen- ja puheenkäsittely, in Finnish) for BSc students.
The Audio Signal Processing Research Group belongs to the Aalto Acoustics Lab. The research group is led by Professor Vesa Välimäki, IEEE Fellow, AES Fellow.
- Artificial reverberation
- Audio filter design (equalizers, delay filters)
- Audio headsets and augmented reality audio
- Deep learning for audio processing
- Digital sound synthesis
- Loudspeaker signal processing
- Time-scale modification of sound
- Virtual analog modeling for music technology
Group members
Latest publications
Multi-shelf graphic equalizer
Efficient Velvet-Noise Convolution in Multicore Processors
Real-time implementation of a linear-phase octave graphic equalizer
Sample rate independent recurrent neural networks for audio effects processing
RIR2FDN: An improved room impulse response analysis and synthesis
Active Acoustics With a Phase Cancelling Modal Reverberator
Differentiable Active Acoustics: Optimizing Stability via Gradient Descent
Non-Exponential Reverberation Modeling Using Dark Velvet Noise
Matching early reflections of simulated and measured RIRs by applying sound-source directivity filters
Sampling the user controls in neural modeling of audio devices
- Published:
- Updated: