- Speech and Audio Signal Processing and Analysis,
- Machine Learning for Audio Signal Processing,
- Biological Signal Processing,
- Speech and Voice Perception,
- Scientific Computing.
- Events detection in Audio signals.
Automatic and reliable detection of pre-defined events in audio signals
is of great importance in many applications such as information
retrieval or smart filtering of audio and has been the subject of
extensive research in recent years.
One such application is automatic detection
of a baby cry in audio signals, which is an essential
step in applications such as remote baby monitoring.
It is also important for researchers, who study the relation
between baby cry patterns and various health or developmental parameters.
In this study we aim to develop efficient machine learning algorithms to
detect the cry events in long and noisy domestic recordings with
high accuracy and very low false positive rate.
Another application is in recording studios,
where the recorded signal may contain unwanted sounds or effects,
and an automatic tool that could detect and manipulate these
sounds is highly desirable.
(with Dima Ruinskiy (Innovation Lab), Rami Cohen (Technion) and Hans IJzerman (University Grenoble Alpes)).
- Bird song recognition and segmentation.
Animal communication and specifically acoustic communication is
in the focus of ecological and biological research.
With the advancement of monitoring technology, a vast amount of
acoustic recordings of birds is continuously accumulated.
As manual segmentation and annotation of this data is impractical, development of efficient algorithms for accurate detection, classification and segmentation of birdsong is therefore a prerequisite for further analysis.
In this study we develop algorithms for automatic
recogntion and segmentation of bird vocalization.
(With Yoni Vortman (Tel-Hai College), Hagai Barmatz (Tel-Aviv University) , Dana Klein (Tel-Hai College) and Sivan Toledo (Tel-Aviv University)).
- Voice Analysis and manipulations.
Voice morphing is the process of producing intermediate or hybrid voices between the utterances of two speakers. It can also be defined as the process of gradually transforming the voice of one speaker to that of another. The ability to change the speakerís individual characteristics and to produce high-quality voices can be used in many applications. Examples include multimedia and video entertainment, as well as enrichment of speech databases in text-to-speech systems.
- Perceptual Learning of Time-Compressed Speech.
(with Karen Banai, Haifa University).
- Speaker Identification, Voice Perception and Modelling.
Human listeners have an extraordinary ability to identify numerous familiar voices, under varying conditions and contexts, in a manner that algorithms for automatic speaker recognition can hardly achieve. Still, little is known about the link between the acoustic features of the speakersí voice and higher processes of speaker identification by the listener. This study aims at examining the relative importance of various acoustic features as cues to familiar speaker identification.