While the state of the art in audio signal processing often assumes a spatially stationary environment, we are in practice often confronted with spatially dynamic sound scenes where moving objects emit or receive sound. This may concern movement of human speakers and listeners, moving audio devices in the context of wireless acoustic sensor networks, autonomous vehicles equipped with microphones or loudspeakers such as robots or drones, or noise emitted by moving vehicles as in road or air traffic. In many processing applications, the movement of acoustic sources or receivers poses challenges, while in others it may be explicitly exploited in that it provides additional spatio-temporal cues as an alternative to discrete spatial sampling.
The wide impact of movement on audio signals is expressed in various spatial, temporal, and spectral aspects. When spatial distances vary, propagation delays become channel- and time-dependent. On the one hand, this obviously affects time-differences of arrival in multi-channel audio signals, which are commonly used to obtain a spatial interpretation of the scene. On the other hand, one also observes spectro-temporal effects, such as frequency-shifts in the well-known Doppler effect. Further, in regards to statistical processing, time-dependent time and frequency shifts question the concept of implementing the expectation operator by averaging over time.
With this Research Topic, we promote the development of movement-aware signal and system models and processing algorithms, that is approaches explicitly addressing the time-variant nature of sound propagation paths caused by the relative movement of acoustic sources and receivers, as well as the spectro-temporal and statistical impact thereof. This covers a wide range of problems including spatial estimation, signal enhancement, system identification, room acoustic information retrieval, and signal generation. Further, in order to enable reproducible and comparative research in the context of movement, we promote the development of relevant databases.
Submissions considered relevant to this Research Topic include, but are not limited to, contributions in the following fields:
Dynamic spatial estimation
- Modeling of movement trajectories
- Acoustic trajectory estimation and location tracking
- Acoustics-based simultaneous localization and mapping (SLAM)
- Echolocation of moving objects
Signal enhancement under movement
- Enhancement of speech, noise reduction and dereverberation
- Movement-robust intrusive performance measures such as quality, intelligibility, and power metrics
Identification of spatially dynamic systems
- Time-varying room impulse response models
- Feedback and echo cancellation
Movement-based room acoustic information retrieval
- Room acoustic feature estimation
- Room acoustic measurement protocols
- Room geometry estimation
Spatially dynamic signal generation
- Movement-aware reverberation models
- Accurate simulation of spatially dynamic acoustic scenes
- Auralization of movement
Acoustic databases involving movement
While the state of the art in audio signal processing often assumes a spatially stationary environment, we are in practice often confronted with spatially dynamic sound scenes where moving objects emit or receive sound. This may concern movement of human speakers and listeners, moving audio devices in the context of wireless acoustic sensor networks, autonomous vehicles equipped with microphones or loudspeakers such as robots or drones, or noise emitted by moving vehicles as in road or air traffic. In many processing applications, the movement of acoustic sources or receivers poses challenges, while in others it may be explicitly exploited in that it provides additional spatio-temporal cues as an alternative to discrete spatial sampling.
The wide impact of movement on audio signals is expressed in various spatial, temporal, and spectral aspects. When spatial distances vary, propagation delays become channel- and time-dependent. On the one hand, this obviously affects time-differences of arrival in multi-channel audio signals, which are commonly used to obtain a spatial interpretation of the scene. On the other hand, one also observes spectro-temporal effects, such as frequency-shifts in the well-known Doppler effect. Further, in regards to statistical processing, time-dependent time and frequency shifts question the concept of implementing the expectation operator by averaging over time.
With this Research Topic, we promote the development of movement-aware signal and system models and processing algorithms, that is approaches explicitly addressing the time-variant nature of sound propagation paths caused by the relative movement of acoustic sources and receivers, as well as the spectro-temporal and statistical impact thereof. This covers a wide range of problems including spatial estimation, signal enhancement, system identification, room acoustic information retrieval, and signal generation. Further, in order to enable reproducible and comparative research in the context of movement, we promote the development of relevant databases.
Submissions considered relevant to this Research Topic include, but are not limited to, contributions in the following fields:
Dynamic spatial estimation
- Modeling of movement trajectories
- Acoustic trajectory estimation and location tracking
- Acoustics-based simultaneous localization and mapping (SLAM)
- Echolocation of moving objects
Signal enhancement under movement
- Enhancement of speech, noise reduction and dereverberation
- Movement-robust intrusive performance measures such as quality, intelligibility, and power metrics
Identification of spatially dynamic systems
- Time-varying room impulse response models
- Feedback and echo cancellation
Movement-based room acoustic information retrieval
- Room acoustic feature estimation
- Room acoustic measurement protocols
- Room geometry estimation
Spatially dynamic signal generation
- Movement-aware reverberation models
- Accurate simulation of spatially dynamic acoustic scenes
- Auralization of movement
Acoustic databases involving movement
