Rolf Bardeli

Rolf Bardeli
Vodafone GmbH · Data Science

Dr. rer. nat.

About

47
Publications
11,968
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
444
Citations
Citations since 2016
3 Research Items
241 Citations
2016201720182019202020212022010203040
2016201720182019202020212022010203040
2016201720182019202020212022010203040
2016201720182019202020212022010203040
Additional affiliations
January 2003 - December 2009
University of Bonn
Education
June 2003 - September 2008
University of Bonn
Field of study
  • Computer Science

Publications

Publications (47)
Conference Paper
In this paper, we enhance and customize the existing BERTopic framework to develop and implement an automated pipeline that delivers a more coherent and diverse set of topics with an even moderate dataset. More specifically, the contributions of this work are threefold: (1) integrate a dynamic and advanced optimizer into the existing BERTopic frame...
Conference Paper
Parkinson's Disease (PD) is a neurodegenerative disorder which gradually effects the neurological condition of the patient. In many cases the disease impairs the reliability of the articulatory system and the ability to pronounce vowels normally. One prominent way to measure the degree of the functioning of the articulatory system is the Vowel Spac...
Chapter
An important task in media synchronisation is to find out the playback position of a running media stream. Only based on this information, it is possible to provide additional information or additional streams synchronised to that running stream. This chapter gives an overview of two techniques for solving this basic task: watermarking and fingerpr...
Article
Event recognition systems have high potential to support crisis management and emergency response. For large-scale scenarios, however, the sheer amount of possible audio and video channels requires adequate processing of the material by automatic means. In this article, the authors focus on automatic audio and video event recognition, by means of d...
Conference Paper
The lack of a formalism for “the problem of music description” results in, among other things: ambiguity in what problem a music description system must address, how it should be evaluated, what criteria define its success, and the paradox that a music description system can reproduce the “ground truth” of a music dataset without attending to the m...
Conference Paper
Evolution and changes of all modern languages is a well-known fact. However, recently it is reaching dynamics never seen before, which results in loss of the vast amount of information encoded in every language. In order to preserve such rich heritage, and to carry out linguistic research, properly annotated recordings of world languages are necess...
Conference Paper
Overlapped speech, where several speakers are speaking simultaneously, is a common occurence in multiparty discussions such as meetings. This kind of speech presents a great challenge to automatic speech processing systems such as speech recognition systems and speaker diarisation systems. In recent speaker diarisation systems, a large portion of t...
Conference Paper
Full-text available
With a constantly rising demand for interactive videos, automatic enrichment of static videos with web links offers seemingly endless possibilities. Given the content that can be found on the web, however, this can be a rather mixed blessing. In this paper, we investigate web sites with contents of extreme physical violence, political extremism and...
Conference Paper
In scenarios with multiple input single output systems, the stochastic constrained least mean-squares (LMS) algorithm has been proven to be an effective approach. However, when only two input channels are available, it is unclear whether this approach still yields improvements. In this paper, we investigate the stableness and the robustness of the...
Chapter
Common approaches to automatic speech recognition (ASR) are based on training statistical models for the acoustics of speech. In our work, a retrieval-based ASR system is developed that does not rely on training and thus provides more flexible application. It is based on a set of known reference word utterances for each possibly occurring word in a...
Conference Paper
Full-text available
With the ever-increasing offer of television content as internet broadcast streams, synchronisation of this material with second screen applications has received considerable interest over the last years. We introduce a novel audio fingerprinting method which can be easily implemented, and offer promising experiments on German news show material. F...
Article
A novel video similarity measure is proposed by using visual features, alignment distances and speech transcripts. First, video files are represented by a sequence of segments each of which contains colour histograms, starting time, and a set of phonemes. After, textual, alignment and visual features are extracted of these segments. The following s...
Conference Paper
Full-text available
The problem of deteriorated automatic speech recognition performance for telecommunication devices like mobile phones is well-studied. However, few papers analyse the influence of emergency broadcast radio devices such as the Terrestrial Trunked Radio (TETRA), which is used for many public safety networks in Europe and Asia. In this paper, we disse...
Article
Full-text available
This review presents our current knowledge on automated methods for acoustic recording of calls and songs of birds. Acoustic long-term recordings can serve as a basis for an automated bird census. We stress the question of whether sound recordings are suitable for qualitative and quantitative analysis of bird populations. Special attention is devot...
Conference Paper
Event recognition systems have high potential to support crisis management and emergency response. Given the vast amount of possible input channels, automatic processing of raw data is crucial. In this paper, we describe several components integrated in an overall intelligent resource management system, namely abnormal event detection in audio and...
Conference Paper
Acoustic modelling is a key issue for successful automatic speech recognition (ASR). Common ASR systems are usually adapted to a certain use case by training robust acoustic models on speech data from the domain recorded in conditions typical for the use case. Varying conditions thus need either multi-conditional or multiple acoustic models. We pre...
Conference Paper
Full-text available
In this work we aim at enriching the transcript of an automatic speech recognition system with punctuation by automatically detecting sentence ends. We make use of a simple word-based language model and combine it with a decision tree for the acoustic features of speech. The focus lies on selecting robust acoustic features that reflect the prosodic...
Article
Full-text available
This review presents our current knowledge on automated methods for acoustic recording of calls and songs of birds. Acoustic long-term recordings can serve as a basis for an automated bird census. We stress the question of whether sound recordings are suitable for qualitative and quantitative analysis of bird populations. Special attention is devot...
Conference Paper
Although abnormal events in an audio stream are by their nature hard to define, a continuous monitoring of audio surveillance data can detect crucial information in, e.g., train engines that might require critical maintenance. Our method detects abnormal events without being trained on a certain situation, by building a model of the expected sound...
Conference Paper
Full-text available
Complex audio scenes with a large number of sound sources pose one of the most difficult problems for audio pattern recognition. Therefore, methods for source separation are very important in this context. Many source separation methods try to exactly recover every source in an audio scene. In this paper, however, we propose an algorithm for the ex...
Article
A large volume of video content on the web is available today, which demands efficient management. To effectively manage, search, retrieve and copy detection, similarity methods play a critical role. In this paper, a novel video similarity measure using visual features, alignment distances and speech transcripts is proposed. Video files are represe...
Article
In the AVATecH project the Max-Planck Institute for Psycholinguistics (MPI) and the Fraunhofer institutes HHI and IAIS aim to significantly speed up the process of creating annotations of audio-visual data for humanities research. For this we integrate state-of-the-art audio and video pattern recognition algorithms into the widely used ELAN annotat...
Article
Trends in bird population sizes are an important indicator in nature conservation but measuring such sizes is a very difficult, labour intensive process. Enormous progress in audio signal processing and pattern recognition in recent years makes it possible to incorporate automated methods into the detection of bird vocalisations. These methods can...
Conference Paper
Full-text available
Typical broadcast material contains not only studio-recorded texts read by trained speakers, but also spontaneous and dialect speech, debates with cross-talk, voice-overs, and on-site reports with difficult acoustic environments. Standard approaches to speech and speaker recognition usually deteriorate under such conditions. This paper reports on t...
Article
AVATecH (Advancing Video/Audio Technology in Humanities Research) is a project in which two Fraunhofer Institutes and two Max Planck Institutes collaborate in order to promote the development and application of technology for semi-automatic annotation of digital audio and video recordings. One of the aims of the AVATecH project is to implement algo...
Conference Paper
Automatic speech recognition for command and control in human-computer interaction (HCI) is used in many applications today. For noisy environments, like the car or motorcycle environment, word error rates of the speech recognition increase and, hence, the reliability of such systems decreases. To achieve an overall robust and reliable system, the...
Article
Full-text available
Systems for speech and speaker recognition already achieve low error rates when applied to high-quality audiovisual broadcast data, such as news shows recorded in a studio environment. Several evaluation corpora exist for this domain in various languages. However, in actual applications for broadcast data analysis, the data requirements are more co...
Article
In the past, similarity search for audio data has largely been focused on music. Recent digitization efforts in some of the larger animal sound archives bring other types of audio recordings into the focus of interest. Although recordings in animal sound archives are usually very well annotated by metadata, it is almost impossible to manually annot...
Thesis
Full-text available
In this thesis, we examine the problem of algorithmic analysis of complex audio scenes with a special emphasis on natural audio scenes. One of the driving goals behind this work is to develop tools for monitoring the presence of animals in areas of interest based on their vocalisations. This task, which often occurs in the evaluation of nature cons...
Article
By definition, a P2-graph $\Gamma$ is an undirected graph in which every vertex is contained in a path of length two. For such a graph, $pc(\Gamma)$ denotes the minimum number of paths of length two that cover all n vertices of $\Gamma$. We prove that $\lceil n/3\rceil\le pc(\Gamma)\le \lfloor n/2\rfloor$ and show that these upper and lower bounds...
Article
In this contribution, we propose a method for the automatic detection and localisation of bird calls featuring simple as well as highly repetitive structures. Reporting from a research project focused at computer aided acoustical monitoring, a detector for Locustella luscinoides (Savi's Warbler) is presented, performing reliably even on highly dist...
Article
Unsupervised recordings of birds for wildlife monitoring usually contain a high number of sources and thus tend to be very complex. This makes automated recognition of bird species much more difficult than from dedicated recordings of individual birds using highly directed microphones. We present studies of bird song recognition algorithms for this...
Book
Full-text available
The animal sound archive of the Humboldt-University is one of the oldest and largest collections of animal sounds. It was founded in 1951 by Günter Tembrock. Presently, the collection consists of about 120,000 bioacoustical recordings comprising almost all groups of animals. Most of the recordings are still on analogue magnetic tape. Currently, the...
Conference Paper
In diesem Beitrag stellen wir eine Projektinitiative zum Aufbau eines verteilten Medienarchivs für bioakustische Datenbestände vor. Im Zuge dieser von der Deutschen Forschungsgemeinschaft (DFG) geförderten Initiative wurde zunächst aufbauend auf den momentan in Digitalisierung befindlichen Datenbeständen des Tierstimmenarchivs der Humboldt Universi...
Conference Paper
Full-text available
Automatic identification of audio titles on radio broadcasts is a first step towards automatic annotation of radio programmes. Systems designed for the purpose of identification have to deal with a variety of postprocessing potentially imposed on audio material at the radio stations. One of the more difficult techniques to be handled is time-scalin...

Network

Cited By

Projects