Contact
Pfaffenwaldring 5 b
70569 Stuttgart
Deutschland
Room: 02.012
Office Hours
Please email me for an appointment.
Subject
AI and Machine Learning
Currently I am working on a project that helps bring Machine Learning to vocational training: KI B3 - Künstliche Intelligenz in die berufliche Bildung bringen. The project is funded by the Federal Ministry for Education and Research, BMBF. In the project I am responsible for creating teaching materials to provide a gentle introduction to AI and Machine Learning for vocational training. The materials for a first course, aimed at students of vocational schools aged roughly 16 to 18, are already completed. The materials are now being extended for further training, then aiming at working professionals. All materials will eventually be available under CC-BY 4.0 license or (in some cases) under CC-BY-SA license.
You can find the current version of the videos in a YouTube Playlist at https://youtube.com/playlist?list=PLqoibb5lIlRRFXSCyTQ_B_Vrl5dcU3zY1
All teaching materials (videos, accompanying PDFs, interactive quizzes, work sheets for group work, Jupyter notebooks) are organized in Moodle courses. Please be aware that all materials are provided in German only. The course for students of vocational schools ("Zusatzqualifikation KI und Maschinelles Lernen") is completed. Its current version 3.3 can be downloaded and imported in your own Moodle. Please see the installation guide (available only in German). If you are planning to use the course, we recommend subscribing to course updates (see at the beginning of the installation document).
Alternatively you can find the contents of this course as individual materials at ZOERR (Zentrales Open Educational Resources Repositorium der baden-württembergischen Hochschulen). All indvidual materials are part of a collection of materials there:
Two more courses for trainings that extend the "Zusatzqualifikation" are being prepared and will be available soon.
Prosody
I am interested in modeling and understanding various aspects of speech and prosody perception and production, with a special interest in the challenges of dealing with spontaneous speech and authentic human-human interactions.
My dissertation thesis was on prosody perception and production in an examplar-theoretic framework. This involved modelling human categorization of prosodic events by applying clustering and machine learning methods to speech data. The parameters that I am particularly interested in are temporal parameters and the PaIntE parameters, which are obtained by parametrizing F0 contours. The classifiers obtained in these experiments have been integrated in a prototype of a tool for automatic prosody prediction.
I have also worked on linguistic text pre-processing for speech synthesis, on prosody prediction and modelling for speech synthesis, and on corpus design for speech synthesis.
Convergence
I was involved in the project Phonetic Convergence in Spontaneous Speech (A4) of the SFB 732. The aim of this project was to investigate convergence of speakers in spontaneous dialogs. Convergence is the process of adapting one's speech to that of an interlocutor. We were interested in understanding exactly which phonetic properties are adjusted in convergence. However, another important aspect of the project was to develop an automatic procedure for creating large databases for phonetic research on spontaneous speech.
Speech databases
I am an expert in the phonetic analysis of large speech databases. To this end I use automatic tools such as forced alignment, parametrization tools, or custom-made Praat and (Linux) shell scripts. I often convert their output into the utterance structure format employed by the Festival speech synthesis toolkit, which allows for integration of phonetic parameters with the linguistic knowledge implemented in Festival, and ultimately for fast extraction of various phonetic parameters including their higher-level linguistic context.
To see a list of my publications with links to most full texts, please check the publications section at the end of this page.
Tutorials
Feel free to download my Praat Scripting Tutorial, example data and solutions.
IMS students might want to also look at my Linux Tutorial or my Unit Selection Tutorial.
(mit freundlicher Erlaubnis von ©TOM)
2023
- Lux, Florian; Koch, Julia; Meyer, Sarina; Bott, Thomas; Schauffler, Nadja; Denisov, Pavel; Schweitzer, Antje; Vu, Ngoc Thang. The IMS Toucan system for the Blizzard Challenge 2023. In: Blizzard Challenge 2023.
- Weigmann, Anja; Knoll, Theresa; Achtenhagen, Claudia; Keppler, Marina; Rott, Karin Julia; Schweitzer, Antje; Ptack, Katharina. Die Gestaltung der Fortbildung "Geprüfte Berufsspezialistin/Geprüfter Berufsspezialist für Künstliche Intelligenz und Maschinelles Lernen (IHK)" im InnoVET-Projekt KI B3 - Künstliche Intelligenz in die berufliche Bildung bringen. In: Wilbers, Karl (editor), Die erste Fortbildungsstufe "Geprüfte/r Berufsspezialist/in der höherqualifizierenden Berufsbildung".
2022
-
Schweitzer, Antje; Möbius, Bernd; Möhler, Gregor; Dogil, Grzegorz. The PaIntE model of intonation. In Jonathan Barnes and Stefanie Shattuck-Hufnagel, editors, Prosodic Theory and Practice MIT Press. Draft PDF
- Sabbatino, Valentino; Troiano, Enrica; Schweitzer, Antje; Klinger, Roman ``splink'' is happy and ``phrouth'' is scary: Emotion Intensity Analysis for Nonsense Words. In Proceedings of the 12th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis pp. 37-50 Association for Computational Linguistics, Dublin, Ireland.
2021
- Schweitzer, Antje. Data-driven prosody modeling using PaIntE. In: Alexander M. Teixeira Kalkhoff; Maria Selig; Christine Mooshammer, Hrsg., Prosody and Conceptional Variation (Studia Romanica et Linguistica), Peter Lang, Frankfurt am Main
2020
- Stehwien, Sabrina; Schweitzer, Antje; Vu, Ngoc Thang. Acoustic and temporal representations in convolutional neural network models of prosodic events. Speech Communication 125:128-141.
2019
-
Schweitzer, Antje. Exemplar-theoretic integration of phonetics and phonology: Detecting prominence categories in phonetic space. Journal of Phonetics 77, pp. 100915. DOI
-
Schweitzer, Antje; Wokurek, Wolfgang; Pützer, Manfred Convergence of Harmonic Voice Quality Parameters in Spontaneous Dialogues. In Proceedings of ICPhS 2019 (Melbourne, Australia), pp. 363-367, 2019. PDF
-
Kügler, Frank; Baumann, Stefan; Andreeva, Bistra; Braun, Bettina; Grice, Martine; Neitsch, Jana; Niebuhr, Oliver; Peters, Jörg; Röhr, Christine T.; Schweitzer, Antje; Wagner, Petra. Annotation of German Intonation: DIMA Compared with other Annotation Systems. In Proceedings of ICPhS 2019 (Melbourne, Australia), pp. 1297-1301, 2019. PDF
-
Jenne, Sabrina; Schweitzer, Antje; Zerbian, Sabine; Vu, Ngoc Thang. Phonological Error Detection for Pronunciation Training Using Neural Spectrogram Recognition. In Proceedings of ICPhS 2019 (Melbourne, Australia), pp. 2961-2965, 2019. PDF
2018
-
Gessinger, Iona; Schweitzer, Antje; Andreeva, Bistra; Raveh, Eran; Möbius, Bernd; Steiner, Ingmar. Convergence of Pitch Accents in a Shadowing Task. In Proceedings of Speech Prosody 2018, Poznan. PDF
-
Stehwien, Sabrina; Vu, Ngoc Thang; Schweitzer, Antje. Effects of Word Embeddings on Neural Network-based Pitch Accent Detection. In Proceedings of Speech Prosody 2018, Poznan. PDF
-
Zellers, Margaret; Schweitzer, Antje. Exploring prosodic and conversational context factors in pitch perception. In Proceedings of Speech Prosody 2018, Poznan. PDF
-
Schweitzer, Katrin; Eckart, Kerstin; Gärtner, Markus; Falenska, Agnieszka; Riester, Arndt; Rösiger, Ina; Schweitzer, Antje; Stehwien, Sabrina; Kuhn, Jonas. German Radio Interviews: The GRAIN Release of the SFB732 Silver Standard Collection. In Proceedings of LREC-2018, Linguistic Resources and Evaluation Conference, Miyazaki.
2017
- Daniel Duran, Natalie Lewandowski, Antje Schweitzer. A 3D computer game for testing perception of acoustic detail in speech. In: POMA - Proceedings of Meetings on Acoustics, vol. 18(1), 2017. http://asa.scitation.org/doi/10.1121/2.0000493
-
Antje Schweitzer, Natalie Lewandowski, Daniel Duran. Social attractiveness in dialogs. In: Proc. of Interspeech, Stockholm, pp. 2243-2247, 2017. PDF
-
Katrin Schweitzer, Michael Walsh, Antje Schweitzer. To see or not to see: Interlocutor visibility and likeability influence convergence in intonation. In: Proc. of Interspeech, Stockholm, pp. 919-923, 2017. PDF
-
Margaret Zellers, Antje Schweitzer. An investigation of pitch matching across adjacent turns in a corpus of spontaneous German. In: Proc. of Interspeech, Stockholm, pp. 2336-2340, 2017. PDF
- Daniel Duran, Natalie Lewandowski, Jagoda Bruni, Antje Schweitzer, "Akustische Korrelate wahrgenommener Persönlichkeitsmerkmale und Stimmattraktivität", in: Jürgen Trouvain, Ingmar Steiner und Bernd Möbius, Hrsg., Elektronische Sprachsignalverarbeitung 2017: Tagungsband der 28. Konferenz, Saarbrücken, TUDpress, pp. 91-98, 2017.
2016
- Antje Schweitzer, Ngoc Thang Vu, "Cross-gender and cross-dialect tone recognition for Vietnamese," Proceedings of Interspeech, pp. 1064-1068, 2016. Full Paper PDF
- Antje Schweitzer, Michael Walsh, "Exemplar dynamics in phonetic convergence of speech rate," Proceedings of Interspeech, pp. 2100-2104, 2016. Full Paper PDF
- Grigorij Aronov, Antje Schweitzer, "Acoustic correlates of word stress in German spontaneous speech," in Tagungsband der 12. Tagung Phonetik und Phonologie im deutschsprachigen Raum (Hrsg. C. Draxler and F. Kleber), (München, Deutschland), pp. 12-15, 2016. Full Paper PDF
- Daniel Duran, Natalie Lewandowski, Antje Schweitzer, "Wahrnehmungsexperimente mit Hilfe eines Computerspiels," in Tagungsband der 12. Tagung Phonetik und Phonologie im deutschsprachigen Raum (Hrsg. C. Draxler and F. Kleber), (München, Deutschland), pp. 39-42, 2016.
- Antje Schweitzer, Natalie Lewandowski, Daniel Duran, "Does phonetic convergence reflect personality?." Abstract and talk presented at the workshop Personality in Speech Production, Cornell, Ithaca, 2016.
- Natalie Lewandowski and Antje Schweitzer, "Social and psychological aspects of phonetic convergence in the geco database". Abstract and talk presented at the Sociolinguistics Symposium 21 (Murcia, Spain), 2016.
- Natalie Lewandowski, Carolin Krämer, Daniel Duran, Antje Schweitzer, "Impact of personality and social factors on phonetic convergence." Abstract and poster presentation at AmLaP (Bilbao, Spain), 2016.
- Daniel Duran, Natalie Lewandowski, Antje Schweitzer: A 3D computer game for testing perception of acoustic detail in speech. Proceedings of 22nd International Congress on Acoustics (ICA); Catholic University of Argentina, Buenos Aires. Paper ICA2016-738. [PDF]
2015
- Antje Schweitzer, Natalie Lewandowski, Daniel Duran, Grzegorz Dogil. Attention, please! - Expanding the GECO database. Proceedings of ICPhS, 2015. - Full Paper PDF (2 MB)
- Frank Kügler, Bernadett Smolibocki, Denis Arnold, Stefan Baumann, Bettina Braun, Martine Grice, Stefanie Jannedy, Jan Michalsky, Oliver Niebuhr, Jörg Peters, Simon Ritter, Christine T. Röhr, Antje Schweitzer, Katrin Schweitzer, Petra Wagner. DIMA - Annotation guidelines for German intonation. Proceedings of ICPhS, 2015. - Full Paper PDF (160 kB)
- Katrin Schweitzer, Michael Walsh, Sasha Calhoun, Hinrich Schütze, Bernd Möbius, Antje Schweitzer, and Grzegorz Dogil. Exploring the relationship between intonation and the lexicon: Evidence for lexicalised storage of intonation. Speech Communication 66, pp. 65-81, Feb. 2015. DOI: http://dx.doi.org/10.1016/j.specom.2014.09.006
- Nadja Schauffler, Antje Schweitzer, Katrin Schweitzer, Petra Augurzky. Avoiding melodic clashes in pitch accent production: a corpus study. In: Leemann, Adrian, Marie-José Kolly, Volker Dellwo, Stephan Schmid (Eds): Trends in phonetics and phonology in German-speaking Europe", 2015.
2014
- Antje Schweitzer, Natalie Lewandowski, and Grzegorz Dogil. Advancing corpus-based analyses of spontaneous speech: Switch to GECO! Poster presentation at LabPhon 2014, July, Tokyo - Abstract PDF
- Antje Schweitzer and Natalie Lewandowski. Social Factors in Convergence of F1 and F2 in Spontaneous Speech. Paper presented at the 10th International Seminar on Speech Production, Cologne. - Abstract PDF (77 kB) - Full Paper PDF (120 kB)
- Natalie Lewandowski and Antje Schweitzer. Degrees of control over influencing factors in phonetic convergence. Abstract presented at the ISSP 2014 Satellite Workshop on “Interpersonal coordination and phonetic convergence”, Cologne.
- Natalie Lewandowski, Antje Schweitzer, Daniel Duran and Grzegorz Dogil An exemplar-based hybrid model of phonetic adaptation. Abstract presented at Georgetown University Round Table on Languages and Linguistics (GURT), Washington, D.C., March 2014. - PDF
2013
- Antje Schweitzer and Natalie Lewandowski. Convergence of Articulation Rate in Spontaneous Speech. Proceedings of Interspeech 2013, Lyon, pp. 525-529, 2013. - PDF (78 kB)
2012
- Sasha Calhoun and Antje Schweitzer. Can intonation contours be lexicalised? Implications for discourse meanings. In Gorka Elordieta Alcibar and Pilar Prieto, editors, Prosody and Meaning (Trends in Linguistics). Mouton DeGruyter, 2012.
- Antje Schweitzer and Natalie Lewandowski. Accommodation of backchannels in spontaneous speech (Abstract). In the booklet of the International Symposium on Imitation and Convergence in Speech, 2012. - PDF (76 kB)
- Antje Schweitzer. Perceptually Motivated Parameters for Automatic Prosodic Annotation (Abstract). Workshop Advancing Prosodic Transcription for Spoken Language Science and Technology, Stuttgart, 2012. - PDF (116 kB)
2011
- Britta Lintfert, Antje Schweitzer, and Bernd Möbius. A parametric approach to intonation acquisition research: Validation on child-directed speech data. In Proceedings of Interspeech 2011, pp. 757-760, 2011. - PDF (214 kB)
- Grzegorz Dogil and Antje Schweitzer. Quantal effects in the Temporal Alignment of Prosodic Events. In Proceedings of the International Congress of Phonetic Sciences (Hong Kong), pp. 595-598, 2011. - PDF (161 kB)
2010
- Antje Schweitzer. Production and Perception of Prosodic Events--Evidence from Corpus-based Experiments. PhD thesis, Universität Stuttgart, 2010. - Catalogue entry - PDF (2.9 MB)
- Natalie Lewandowski and Antje Schweitzer. Prosodic and segmental convergence in spontaeous German conversations (A). Journal of the Acoust. Society of America 128(4), p. 2458, 2010.
- Britta Lintfert, Antje Schweitzer, Lukasz Wolski, and Bernd Möbius. Quantifying developmental changes of prosodic categories. In Speech Prosody 2010 (Chicago, USA). - PDF (127 kB)
- Katrin Schweitzer, Sasha Calhoun, Hinrich Schütze, Antje Schweitzer, and Michael Walsh. Relative frequency affects pitch accent realisation: Evidence for exemplar storage of prosody. In Proceedings of the 13th Australasian International Conference on Speech Science and Technology (Melbourne), 2010. - PDF (177 kB)
2009
- Antje Schweitzer and Bernd Möbius. Experiments on automatic prosodic labeling. In Proceedings of Interspeech 2009, pp. 2515-2518, 2009. - PDF (119 kB)
- Katrin Schweitzer, Michael Walsh, Bernd Möbius, Arndt Riester, Antje Schweitzer, and Hinrich Schütze. Frequency matters: Pitch accents and information status. In Proceedings of EACL-09 (Athens, Greece), pp. 728-736, 2009. - PDF (311 kB)
- Jagoda Sieczkowska, Bernd Möbius, Antje Schweitzer, Michael Walsh, Grzegorz Dogil. Voicing profile of Polish consonants: [r] in obstruent clusters. In Proceedings of Interspeech 2009 (Brighton) pp. 2319-2322, 2009. - PDF (311 kB)
2007
- Martin Barbisch, Bettina Säuberlich, and Antje Schweitzer. Zwei hybride Unit Selection-Strategien im Vergleich. In Fortschritte der Akustik--DAGA 2007 (Stuttgart), 2007a
- Martin Barbisch, Grzegorz Dogil, Bernd Möbius, Bettina Säuberlich, and Antje Schweitzer. Unit selection synthesis in the SmartWeb project. In Proceedings of the Sixth ISCA Workshop on Speech Synthesis (SW6-2007), pp. 304-309, 2007b. - PDF (329 kB)
- Michael Walsh, Hinrich Schütze, Bernd Möbius, and Antje Schweitzer. An exemplar-theoretic account of syllable frequency effects. In Proceedings of the International Congress of Phonetic Sciences (Saarbrücken), pp. 481-484, 2007.
2006
- Antje Schweitzer, Norbert Braunschweiler, Grzegorz Dogil, Tanja Klankert, Bernd Möbius, Gregor Möhler, Edmilson Morais, Bettina Säuberlich, and Matthias Thomae. Multimodal speech synthesis. In Wolfgang Wahlster, editor, SmartKom: Foundations of Multimodal Dialogue Systems, pp. 411-435. Springer, 2006. Draft PDF
2004
- Antje Schweitzer and Bernd Möbius. Exemplar-based production of prosody: Evidence from segment and syllable durations. In Speech Prosody 2004 (Nara, Japan), pp. 459-462, 2004. - PDF (169 kB)
- Antje Schweitzer, Norbert Braunschweiler, Grzegorz Dogil, and Bernd Möbius. Assessing the acceptability of the SmartKom speech synthesis voices. In Proceedings of the 5th ISCA Speech Synthesis Workshop (Pittsburgh, PA), pp. 1-6, 2004. - PDF (169 kB)
2003
- Antje Schweitzer, Norbert BraunschweiDraft PDFler, Tanja Klankert, Bernd Möbius, and Bettina Säuberlich. Restricted Unlimited Domain Synthesis. In: Proceedings of Eurospeech 2003 (Geneva), vol. 2, pp. 1321-1324, 2003. - PS (1146 kB)
- Antje Schweitzer and Bernd Möbius. On the structure of internal prosodic models. In Proceedings of the 15th International Congress of Phonetic Sciences (Barcelona), pp. 1301-1304, 2003. - PS (182 kB)
- Antje Schweitzer and Bernd Möbius. Temporal constraints on the production of frequent and infrequent syllables. Abstract presented at the 6th International Seminar on Speech Production (Sydney), 2003.
2002
- Antje Schweitzer, Norbert Braunschweiler, and Edmilson Morais. Prosody Generation in the SmartKom Project. In: Bernard Bel and Isabel Marlien (eds.), Proceedings of Speech Prosody 2002 (Aix-en-Provence, Laboratoire Parole et Langage), 639-642, 2002. - PDF (30 kB)
2001
- Anton Batliner, Bernd Möbius, Gregor Möhler, Antje Schweitzer, Elmar Nöth. Prosodic models, automatic speech understanding, and speech synthesis: towards the common ground. Proceedings of the European Conference on Speech Communication and Technology (Aalborg, Denmark), vol. 4, 2285-2288, 2001. - PDF (40 KB)
- Antje Schweitzer. Gesture-speech interaction in the SmartKom project (A). Journal of the Acoustical Society of America, 110(5):2705-2706, 2001. - A4-sized handout, PDF (194 kB)
2000
- Antje Schweitzer and Martin Haase. Zwei Ansätze zur syntaxgesteuerten Prosodiegenerierung. In Konvens-2000 (Ilmenau, Germany), pp. 197-202, 2000. - PDF (162 kB)
- Bernd Möbius, Gregor Möhler, Antje Schweitzer, Anton Batliner, and Elmar Nöth. Prosodic models and speech synthesis: towards the common ground. In Stanislaw Puppel and Grazyna Demenko, editors, Prosody 2000: Speech Recognition and Synthesis (Kraków, Poland), pp. 155-160, Poznan, Poland, 2000.