Room number

IDEG110

Telephone number

• office: +43 316 873 - 4433

Position

Research and Teaching Associate

Email

linke@tugraz.at

Julian Linke (Dipl.-Ing. Dr.techn. BSc) is an interdisciplinary researcher since his Master’s studies, combining the fields of acoustics, psychoacoustics, speech technology and speech science. His work aims at increasing our understanding of everyday communication. In more detail, he focuses on the development of automatic speech recognition systems and automatic tools for supra-segmental analyses, on the automatic detection of vocal fatigue, and on the integration of linguistic knowledge into speech technology. He further has substantial experience with data management, as well as experience with recording technology for the development of speech corpora.

Links: LinkedIn, Google Scholar.

---

Research Projects

• TUNE-IN
• Cross-layer prosody models for conversational speech (FWF Elise Richter Grant V638-N33)

Courses

• Digital Signal Processing Laboratory
• Signal Processing
• Signaltransformationen

Student Projects

• Development of a transcription system for the documentation of endangered languages(open)
• Integration of prosodic features to ASR systems
• Automatic Pitch Accent Annotation of Conversational German
• Integration of Prosodic Features to Automatic Speech Recognition Systems
• Automatic Speech Recognition on the Vienna Scientific Cluster(open)
• Applying Semi-Supervised Learning to Adventitious Lung Sound Classification
• Analysis of Acoustic Parameters of Vowels for Detecting Voice Changes in Long-Term Voice Recordings
• Analysis of Acoustic Parameters for Detecting Voice Changes in Long-Term Voice Recordings

PhD Thesis

• What's so complex about conversational speech? Prosodic prominence and speech recognition challenges

Publications

• Conference paper Linke J., Steger S., Steinwender P., Kubin G., Pernkopf F. & Schuppler B. (2025) Uncertainty prediction for prominence classification with chroma features. in 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025 (pp. 1). [more info] [doi]
• Journal article Linke J., Geiger B., Kubin G. & Schuppler B. (2025) What’s so complex about conversational speech? A comparison of HMM-based and transformer-based ASR architectures. in Computer Speech and Language , 90. [more info] [doi]
• Conference paper Karner M., Linke J., Kroell M., Schuppler B. & Geiger B. (2024) Towards Improving ASR Outputs of Spontaneous Speech with LLMs. in 20th Conference on Natural Language Processing, KONVENS 2024 (pp. 339-348). [more info]
• Conference paper Mihajlik P., Meng Y., S. M., Linke J., Schuppler B. & Mady K. (2024) On Disfluency and Non-lexical Sound Labeling for End-to-end Automatic Speech Recognition. in 25th Annual Conference of the International Speech Communication Association (pp. 1270-1274). [more info] [doi]
• Journal article Pokorny F., Linke J., Seddiki N., Lohrmann S., Gerstenberger C., Haspl K., Feiner M., Hagmüller M., Schuppler B., Kubin G., Gugatschka M. & Eyben F. (2024) VocDoc, what happened to my voice? Towards automatically capturing vocal fatigue in the wild. in Biomedical Signal Processing and Control, 88(B). [more info] [doi]
• Conference paper Hagmüller M., Linke J., Lohrmann S., Pokorny F. & Schuppler B. (2023) An acoustic analysis of vowels to predict voice changes in a long reading task. in 13th International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (pp. 23-26). [more info] [doi]
• Conference paper Mihajlik P., Kadar M., Dosinszky G., Meng Y., Kedalai M., Linke J., Fegyo T. & Mady K. (2023) What kind of multi- or cross-lingual pre-training is the most effective for a spontaneous, less-resourced ASR task?. in 2nd Annual Meeting of the Special Interest Group on Under-resourced Languages: SIGUL 2023. [more info]
• Conference paper Linke J., Kadar M., Dosinszky G., Mihajlik P., Kubin G. & Schuppler B. (2023) What do self-supervised speech representations encode? An analysis of languages, varieties, speaking styles and speakers. in 24th Annual Conference of the International Speech Communication Association (pp. 5371-5375). [more info] [doi]
• Preprint Linke J., Wepner S., Kubin G. & Schuppler B. (2023) Using Kaldi for Automatic Speech Recognition of Conversational Austrian German.. [more info] [doi]
• Conference paper Linke J., Kubin G. & Schuppler B. (2023) Using word-level features for prosodic prominence detection in conversational speech. in 20th International Congress on Phonetic Sciences (pp. 3101). [more info]
• Conference paper Kelterer A., Wepner S., Linke J. & Schuppler B. (2023) Points of maximum grammatical control – The prosody of a turn-holding practice. in 20th International Congress on Phonetic Sciences (pp. 3467-3471). [more info]
• Conference paper Linke J., Garner P., Kubin G. & Schuppler B. (2022) Conversational Speech Recognition Needs Data? Experiments with Austrian German. (pp. 4684–4691). [more info]
• Abstract Kelterer A., Christian S., Wepner S., Linke J. & Zarka D. (2021) Prosodic cues to agreement and disagreement in "ja" and "nein" prefaces in Austrian German conversations.. [more info]
• Conference paper Linke J., Kelterer A., Dabrowski M., Zarka D. & Schuppler B. (2020) Towards automatic annotation of prosodic prominence levels in Austrian German. in 10th International Conference on Speech Prosody (pp. 1000 - 1004). [more info] [doi]