Felix Burkhardt

Bio: Felix Burkhardt is an academic researcher from University of Freiburg. The author has contributed to research in topics: Medicine & Speech synthesis. The author has an hindex of 24, co-authored 87 publications receiving 4031 citations. Previous affiliations of Felix Burkhardt include Humboldt University of Berlin & Technical University of Berlin.

A database of German emotional speech.

Abstract: The article describes a database of emotional speech. Ten actors (5 female and 5 male) simulated the emotions, producing 10 German utterances (5 short and 5 longer sentences) which could be used in everyday communication and are interpretable in all applied emotions. The recordings were taken in an anechoic chamber with high-quality recording equipment. In addition to the sound electro-glottograms were recorded. The speech material comprises about 800 sentences (seven emotions * ten actors * ten sentences + some second versions). The complete database was evaluated in a perception test regarding the recognisability of emotions and their naturalness. Utterances recognised better than 80% and judged as natural by more than 60% of the listeners were phonetically labelled in a narrow transcription with special markers for voice-quality, phonatory and articulatory settings and articulatory features. The database can be accessed by the public via the internet (http://www.expressive-speech.net/emodb/).

The INTERSPEECH 2010 Paralinguistic Challenge

Abstract: Most paralinguistic analysis tasks are lacking agreed-uponevaluation procedures and comparability, in contrast to more‘traditional’ disciplines in speech analysis. The INTERSPEECH2010 Paralinguistic Challenge shall help overcome the usuallylow compatibility of results, by addressing three selected sub-challenges. In the Age Sub-Challenge, the age of speakers hasto be determined in four groups. In the Gender Sub-Challenge,a three-class classification task has to be solved and finally, theAffect Sub-Challenge asks for speakers’ interest in ordinal rep-resentation. This paper introduces the conditions, the Challengecorpora “aGender” and “TUM AVIC” and standard feature setsthat may be used. Further, baseline results are given.Index Terms: Paralinguistic Challenge, Age, Gender, Affect 1. Introduction Most paralinguistic analysis tasks resemble each other not onlyby means of processing and ever-present data sparseness, but bylacking agreed-upon evaluation procedures and comparability,in contrast to more traditional disciplines in speech analysis. Atthe same time, this is a rapidly emerging field of research, dueto the constantly growing interest on applications in the fieldsof Human-Machine Communication, Human-Robot Communi-cation, and Multimedia Retrieval. In these respects, the INTER-SPEECH 2010 Paralinguistic Challenge shall help bridging thegap between excellent research on paralinguistic information inspoken language and low compatibility of results, by address-ing three selected tasks. The “aGender” and the “TUM AVIC”corpora are provided by the organizers. The first consists of 46hours of telephone speech, stemming from 954 speakers, andserves to evaluate features and algorithms for the detection ofspeaker age and gender. The second features 2 hours of humanconversational speech recording (21 subjects), annotated in 5differentlevelsofinterest. Thecorpusfurtherfeaturesauniquelydetailed transcription of spoken content with word boundaries byforced alignment, non-linguistic vocalizations, single annotatortracks, and the sequence of (sub-)speaker-turns. Both are given

The INTERSPEECH 2012 Speaker Trait Challenge

Abstract: Keywords: Computational Paralinguistics ; Speaker Traits ; Personality ; Likability ; Pathology Reference EPFL-CONF-174360 Record created on 2012-01-23, modified on 2017-05-10

Verification of acoustical correlates of emotional speech using formant-synthesis

Abstract: This paper explores the perceptual relevance of acoustical correlates of emotional speech by means of speech synthesis. Besides, the research aims at the development of emotion-rules which enable an optimized speech synthesis system to generate emotional speech. Two invesigations using this synthetizer are described : 1) the systematic variation of selected acoustical features to gain a preliminary impression regarding the importance of certain acoustical features for emotional expression, and 2) the specific manipulation of a stimulus spoken under emotionally neutral condition to investigate further the effect of certain features and the overall ability of the synthetizer to generate recognizable emotional expression. It is shown that this approach is indeed capable of generating emotional speech that is recognized almost as well as utterances realized by actors

A database of German emotional speech.

Abstract: The article describes a database of emotional speech. Ten actors (5 female and 5 male) simulated the emotions, producing 10 German utterances (5 short and 5 longer sentences) which could be used in everyday communication and are interpretable in all applied emotions. The recordings were taken in an anechoic chamber with high-quality recording equipment. In addition to the sound electro-glottograms were recorded. The speech material comprises about 800 sentences (seven emotions * ten actors * ten sentences + some second versions). The complete database was evaluated in a perception test regarding the recognisability of emotions and their naturalness. Utterances recognised better than 80% and judged as natural by more than 60% of the listeners were phonetically labelled in a narrow transcription with special markers for voice-quality, phonatory and articulatory settings and articulatory features. The database can be accessed by the public via the internet (http://www.expressive-speech.net/emodb/).

Communication of emotions in vocal expression and music performance: different channels, same code?

Abstract: Many authors have speculated about a close relationship between vocal expression of emotions and musical expression of emotions. but evidence bearing on this relationship has unfortunately been lacking. This review of 104 studies of vocal expression and 41 studies of music performance reveals similarities between the 2 channels concerning (a) the accuracy with which discrete emotions were communicated to listeners and (b) the emotion-specific patterns of acoustic cues used to communicate each emotion. The patterns are generally consistent with K. R. Scherer's (1986) theoretical predictions. The results can explain why music is perceived as expressive of emotion, and they are consistent with an evolutionary perspective on vocal expression of emotions. Discussion focuses on theoretical accounts and directions for future research.