Student Projects
- Analyzing the Cost of Training Physics-Informed Neural Networks (PINNs)(open)
- Effect of a Hybrid Meeting Situation on Communication Quality(open)
- Automatic Speech Recognition for Dementia Prediction(open)
- Voice Conversion on Embedded Devices(open)
- Streaming ASR for Pathological and Alaryngeal Speech(open)
- Backchannel Oppertunity Points with a Social Robot(open)
- Sicherheit von Netzgekoppelten Batterie Großspeicher Anlagen(open)
- Simulation of Standalone Battery Energy Storage Systems in Electricity Markets(open)
- AI-supported Fault Detection in PV Systems Portfolio(open)
- Development of a Digital Twin for PV Systems with Real-Time Data Connection(open)
- Automated AI-Based Thermography Analysis and Maintenance for PV Systems(open)
- Deep Neural Network Models of Auditory Processing in the Human Brain(open)
- Automatic Event Classification for Distributed Acoustic Sensing (DAS)(open)
- Accepted latency for voice conversion speaking aid.(open)
- Projects with the Intelligent Systems Group(open)
- Streaming voice conversion for the improvement of pathological speech(open)
- Language and pronunciation assessment in speech-language pathologies(open)
- Audio-visual speech enhancement for the improvement of pathological speech(open)
- Voice conversion for disease progression modelling(open)
- Prosthetics for hearing & cognitive impairments(open)
- Deep clustering of pathological speakers(open)
- Breathing patterns in text readings of pathological speakers(open)
- Sonification of Electric Power Grid Data(open)
- Simulation of Auditory Masking Tests for Outer Hair Cell Model(open)
- Sequential Multi-pitch Tracking(open)
- Hybrid ML for Music Analysis(open)
- Development of a transcription system for the documentation of endangered languages(open)
- Abschätzung der Empfangsqualität von Mobilfunk und Power Line Communication für Smart Meter(open)
- Design and Implementation of an Navigation System for Visually Impaired based on Sonification of 3D Scene Measurements(open)
- Design and Implementation of an Navigation System for Visually Impaired based on Sonification of 3D Scene Measurements(open)
- Design and Implementation of an Navigation System for Visually Impaired based on Sonification of 3D Scene Measurements(open)
- Bird Scanner(open)
- Bionic Voice Projects(open)
- Effects of Room Acoustics on Conversational Speech
- Furhat -- Interaction with a Social Robot(open)
- Assistive tool for a child with dysarhric speech and limited fine motor skills(open)
- Classification of turn-taking and conversational dynamics(open)
- Automatic Pitch Accent Annotation of Conversational German
- Classification of Communicative Functions(open)
- Miniaturlautsprecher für die Verwendung in Haushaltsgeräten (Juni 2023)
- Room acoustic adaptation of a living room for rehearsal purposes / Raumakustische Adaptierung eines Wohnraums für Übungszwecke
- Anomaly Detection in Industrial Applications(open)
- Acoustic Event Detection and Classification(open)
- Generative Deep Learning for the Creation of 3D Objects(open)
- Active Noise Cancellation Demo
- Self-explaining recording device(open)
- Add features to our Bionic Voice on a Raspberry Pi(open)
- Automatic Speech Recognition on the Vienna Scientific Cluster(open)
- Adaption der Lightmap als Datenstruktur in der raumakustischen Simulation (Dezember 2022)
- Neural-Enhanced Multipath-based SLAM(open)
- Comparison of Automatic and Human Speech Recognition
- Alignment of Laser Measurements for Refractory Modelling
- Applying Semi-Supervised Learning to Adventitious Lung Sound Classification
- Improving Probabilistic Inference(open)
- Speaker Diarization and Recognition for RadioPlays with SpeechBrain(open)
- Machine Learning for Automatic F0 Contour Discrimination and Labelling
- Automatic Labelling of Creaky Patterns in Speech Signals
- Speaker Diarization using a small microphone array(open)
- Monitoring of Dairy Cows with Deep Learning
- Anomaly Detection in Dairy Cows
- Room acoustic planning for the new building of the social association Alphanova / Raumakustische Planung für das neue Gebäude des Sozialvereins Alphanova
- Machine Learning for Discharge Prediction in Nuclear Fusion Reactors
- Architectural Acoustics - Evaluation of Impulse Responses in Concert Halls(open)
- Requirements for room acoustics in the context of video telephony / Anforderungen an die Raumakustik im Kontext von Videotelefonie
- Vibration Diagnosis in Automotive Applications
- Akustik für Motor und Fahrzeug
- Homophone disambiguation for conversational Austrian German(open)
- Generative adversarial networks for time-series tasks
- Raumakustische Sanierung eines Mehrzweckraumes / Room acoustic renovation of a multi-purpose room
- Keyword spotting using resource-efficient deep learning
- Room Acoustic Optimization of a Multi-purpose Room / Raumakustische Untersuchung eines Mehrzweckraums
- Improving inference runtime for Dirichlet Process Bayesian Neural Networks
- Sound Level Monitoring Tool
- Bayesian Neural Networks for Missing Data Scenarios
- Resource-efficient Neural Networks
- Oxygen Measurements for Apnea Divers
- Room Acoustic Investigation of Robot Laboratories / Raumakustische Untersuchung von Roboter Laboren
- Microphone Arrays for Lecture Halls
- Raumakustische- und elektroakustische Planung für Ambisonics Produktionsräume
- Raumakustische Sanierung von Räumen der NMS Vorau
- Training Tool for Speaking with a Bionic Voice(open)
- Synthesis of disordered voices(open)
- Bionic Voice Intelligibility Test(open)
- Weiterentwicklung eines FPGA Laborplatzes für digital Audio(open)
- Parameters for gender adequate voice perception(open)
- Probabilistic Graphical Models for Signal Analysis and Recognition
- Architectural Acoustics - Acoustic Design of Lecture Halls
- Automatic Transcription of Analog Space Missions(open)
- Space/Mars Mission conversation analysis(open)
- Richtungsbeschallung und virtuelle Raumakustik am Beispiel der Opernproduktion "Maria de Buenos Aires"
- Exploration of unknown terrain(open)
- An adaptive filter based on recursive blockmatrix inversion(open)
- Akustische Bestandsaufnahme der Hörsäle der TU Graz
- Dokumentation einer raumakustischen Sanierung mit Hörbeispielen(open)
- Positionsbestimmung in der Ventiltechnik (Fa. Ventrex)(open)
- Tracking of fundamental frequencies in pathologic voice(open)
- Optimierung eines akustischen Prüfstandes (November 2018)
- Singing with a Bionic Voice(open)
- Room acoustic design of the day center Robert Stolz - Raumakustische Planung im Tageszentrum Robert Stolz
- Multi E-Larynx Stimulation
- Bionic Voice Design(open)
- Realtime Digital Signal Processing using Raspberry Pi(open)
- Change Point Detection in Smartphone Usage
- Ultrasonic Inspection System(open)
- Stresserkennung aus Sprachdaten(open)
- Mobile Multichannel Playback System
- Loudness balance control in hear-through systems(open)
- Evaluierung von Cloud Speech APIs(open)
- Gradient flow localizer with a miniature microphone array(open)
- Raumakustische Messungen mit Apps
- Planung und Evaluierung von Kantenabsorbern
- Measurement of Impact of Pop Sound Distortions on Automatic Speech Recgonition(open)
- Raumakustische Optimierung eines Aufnahmeraumes im Tonstudio der Kunstuniversität Graz
- Entwicklung und Realisierung von kleinen Schaltungen(open)
- Pop Sound Detection and Suppression(open)
- Pop Informed Speech Recognition(open)
- Implementation of a modular audio analysis software in Python
- Robust Distance Estimation for Time-of-Flight 3D Image Sensors(open)
- Localization and Synchronization(open)
- Machine Translation with Recurrent Neural Networks(open)
- Jointly Model Source Separation and Speech Recognition(open)
- Voice-controlled home automation
- Building a Prototype of Multipath-assisted Localization(open)
- Implementation of a Maximum Negentropy Beamformer(open)
- Implementation of a Maximum Kurtosis Beamformer(open)
- EL Speech improvement on a Raspberry Pi(open)
- Vibration Analysis and Classification(open)
- PAK-Messablaufsteuerungen für Laborübungen der akustischen Messtechnik (Jänner 2016)
- IFEN: Signal Processing for Satellite Navigation Systems(open)
- Speaker Recognition with deep neural networks(open)
- Single Channel Source Separation with Deep Model: Tackle the Chime3 challenge(open)
- Reduction of Heart Sounds from Lung Sounds(open)
- Machine Translation with Deep Neural Networks
- Development and Comparison of Lung Sound Transducers(open)
- Acoustic Imaging of the Human Respiratory System(open)
- Subjective Measures for Electro-Larynx Speech(open)
- Parametrization of a Transducer for EL Speech Production II(open)
- Wake-up-word recognition
- Modelling of gas sensor responses using deep models(open)
- Impact of signal pre-processing on deep model complexity(open)
- BatMAN: Bat-Like Multipath-assisted Navigation(open)
- High-Accuracy Indoor Localization Exploiting Multipath(open)
- Automatic Speech Recognition for Austrian German on a Raspberry Pi
- Accurate Distance Estimation for RFID Positioning(open)
- Semi(supervised) speech enhancement using Non-negative matrix factorization(open)
- Psychoakustischer Rauschmodulator(open)
- Roomacoustic Optimizations for Winter Gardens; Raumakustische Optimierungen für Wintergärten
- Akustische Sanierung am Bundesrealgymnasium Kepler
- Automatic Speech Recognition for Electro-Laynx Speech
- LED-Stroboskop/Signalgenerator für Lautsprechermessungen
- Vermessung und Erarbeitung von Optimierungsvorschlägen für die Raumakustik mehrerer Räume im Palais Herberstein in Graz
- Implementierung eines Beschallungssystems bestehend aus 4 nachführbaren parametrischen Lautsprecherarrays(open)
- Making Wireless Communications Visible: Implementation of Transmission Demos using a Channel Sounder
- Design/Implementation of digital baseband receiver algorithms(open)
- Using sEMG for Disordered Speech Enhancement II
- Reference-Free Speech Quality Measure for Electro-Larynx Speech(open)
- Topics in Markov Chains II(open)
- Topics in Markov Chains I(open)
- Implementation of an audio processing algorithm as gstreamer plugin(open)
- Artificial Pitch for Electro-Larynx Speech(open)
- Overlapping talk and its relationship to turn-taking(open)
- Pop Sound Protection and Reduction for Small Microphone Enclosures
- Acoustically Coupled Microphones(open)
- Real-time Controllable Sound Source Generator for Electro-Larynx Speech(open)
- KLIPPEL Messplatz(open)
- Master Theses at FTW(open)
- Single-Channel Source Separation in Multisource Environment(open)
- Sample Based Glottal Excitation Signal Database(open)
- Realtime Model-based Single-Channel Speech Separation(open)
- Realisierung eines motorisierten Gitarrenstimmgerätes
- Exact Maximum Margin Structure Learning(open)
- Burst Detection in Plosives for ASR
- Analyzing the Probabilities of GPS Acquisition(open)
- Kalibrierung von pv-Sonden (Juni 2013)
- Bestimmung des Luftschallleistungspegels mittels Vibrationsmessung (April 2013)
- Master Theses at ABB, Sweden(open)
- Master Theses at INTEL, Linz(open)
- Structure Learning of Bayesian Networks Using Latent Variables(open)
- Ventilierte Lautsprecherbox mit Koaxialchassis(open)
- Helmholtz-Absorber(open)
- Akustikworkshop für die JuniorUni Graz(open)
- Weighted Maximum Margin Bayesian Network Classifiers(open)
- Analogfilter für Audio
- Dictionary-based Speech Enhancement using an Air-Bone-Microphone
- Charakterisierung der Schädigung von Hüfttotalendoprothesen mittels Körperschall(open)
- Automatic Differentiation for Large Scale Optimization(open)
- Large Scale Neural Networks for Audio Source Separation(open)
- Eigenfrequenzen - Eigenmoden, Resonanzfrequenzen - Raummoden
- Anwendung von Funkvernetzungen in der Beschallungstechnik
- Hybrid Generative-Discriminative Classification Models for Semi-Supervised Learning in Speech Applications(open)
- Floating-Point Matrix Multiplication on FPGAs(open)
- Verstehen, Finden und Eliminieren von Erdschleifen in Audiosystemen
- Auralisation of the Herz Jesu Church in Graz
- Lautsprecherentwicklung für das Elektroakustik Labor
- Measurement Setup for acoustic two-port Networks
- Comparison of Excitation Signals for an Electronic Larynx
- Messsystem für akustische Zweitore (April 2008)
- A Flexible Sender-Based Packet Loss Recovery Method
- Theses & Projects
PhD Theses
- PhD Theses
- The prosody of interactional and discursive strategies in Austrian conversational speech
- A Holistic Approach to Multi-channel Lung Sound Classification
- Adaptive Calibration of Frequency Response MIsmatches in Time-Interleaved Analog-to-Digital Converters
- Adaptive Digital Predistortion of Nonlinear Systems
- Sum-Product Networks for Complex Modelling Scenarios
- Audiovisual Speech Synthesis Based on Hidden Markov Models
- Automated Anomaly Classification for the Post-Mortem System using Machine Learning
- Bayesian Causal Inference in the Presence of Structural Uncertainty
- Complex Baseband Modeling and Digital Predistortion for Wideband RF Power Amplifiers
- What's so complex about conversational speech? Prosodic prominence and speech recognition challenges
- Generalized Noncoherent Ultra-Wideband Receivers
- High-Accuracy Positioning Exploiting Multipath for Reducing the Infrastructure
- Improving Efficiency and Generalization in Deep Learning Models for Industrial Applications
- PhD Theses
- Information Loss in Deterministic Systems Connecting Information Theory, Systems Theory, and Signal Processing
- Information Loss in Deterministic Systems - Connecting Information Theory, Systems Theory, and Signal Processing
- Low Complexity Correction Structures for Time-Varying Systems
- Low Complexity Ultra-wideband (UWB) Communication Systems in Presence of Multiple-Access Interference
- Machine learning based scene representation and condition monitoring
- Maximum Margin Bayesian Networks
- Modeling and Mitigation of Narrowband Interference for Non-Coherent UWB Systems
- Modeling, Identification, and Compensation of Channel Mismatch Errors in Time-Interleaved Analog-to-Digital Converters
- Modelling and simulation of porous absorbers in room edges
- Multipath-Assisted Indoor Positioning
- Multipath Tracking and Prediction for Multiple Input Multiple Output Wireless Channels
- Nonlinear System Identification for Mixed Signal Processing
- Understanding the Behavior of Belief Propagation
- Analysis of Message Passing Algorithms and Free Energy Approximations in Probabilistic Graphical Models
- Physics-informed Machine Learning
- (When) Does it Harm to Be Incomplete? Comparing Human and Automatic Speech Recognition of Syntactically Disfluent Utterances
- Psychoacoustic Modelling of Selective Listening in Music
- Robust Test-Time Adaptation for Visual and Multimodal Learning under Distribution Shifts
- Signal Processing for Energy-Efficient Burst-Mode RF Transmitters
- Signal Processing for Ultra Wideband Transceivers
- Signal Processing in Phase-Domain All-Digital Phase-Locked Loops
- Social robots for training medical conversations
- Adaptive Digital Predistortion of Nonlinear Systems
- Uncertainty Estimation in Deep Learning and Industrial Applications
- Voice conversion for Dysphonic and Electrolaryngeal Speech