Posters
Updated Best Student Poster Award
Robustness of biologically motivated features for DNN-based ASR
Fernando de la Calle, Universidad Carlos III de Madrid
All posters will be available from the first till the last day of the
Summer School to facilitate the discussion between participants.
Poster sessions are divided into two parts: a session
of 15 minutes where students have a spotlight presentation of 2 min
each in the lectures room, and a formal poster presentation of 45
min in the posters room
Poster size: A0, portrait orientation (in fact, landscape is also possible)
The Best Student Poster Awards will recognize the most outstanding student poster presented during RTTHSS2015. It will be given on Wednesday July 8, during the Summer School Dinner.
Poster Session 1: Tuesday July 7
P1-1 | Prosodic break prediction with sintactic features |
Santi Pascual, TALP, Universitat Politècnica de Catalunya | |
P1-2 | Is it Harry? Or Voldemort? Is he sad or angry? - New approaches to expressive speech synthesis |
Igor Jauk, TALP, Universitat Politècnica de Catalunya | |
P1-3 | Generating Singing Voice Expression Contours Based on Unit Selection |
Martí Umbert, MTG, Universitat Pompeu Fabra | |
P1-4 | Singing voice conversion using generative layer-wise trained DNN |
Merlijn Blaauw, MTG, Universitat Pompeu Fabra | |
P1-5 | The Use of Voice-quality Measurements in Speaker Diarization |
Abraham Woubie, TALP, Universitat Politècnica de Catalunya | |
P1-6 | The speech analogy to music transcription in syllabic percussion systems |
Ajay Srinivasamurthy, MTG, Universitat Pompeu Fabra | |
P1-7 | Deep Neural Network Feature Classification for Speaker Verification |
Pooyan Safari, TALP, Universitat Politècnica de Catalunya | |
P1-8 | The Role of Emotions, Prosody and Voice Quality in Storytelling Speech across Languages |
Raúl Montaño, La Salle, Universitat Ramon Llull | |
P1-9 | Using Lyrical Duration for Automatic Lyrics-to-Audio Alignment in Classical Turkish Music |
Georgi Dzhambazov, MTG, Universitat Pompeu Fabra |
Poster Session 2: Wednesday July 8
P2-1 | Noise-Robust ASR on Portable Electronic Devices with Small Microphone Array |
Iván López, Universidad de Granada | |
P2-2 | Multimodal Combination of Multiple Handwriting and Speech Recognition Systems |
Emilio Granell, Universitat Politècnica de València | |
P2-3 | Automatic Speech Recognition for Tigrigna Using Deep LSTM Approach |
Hafte Abera, Addis Abeba University, Ethiopia | |
P2-4 | Robustness of biologically motivated features for DNN-based ASR |
Fernando de la Calle, Universidad Carlos III de Madrid | |
P2-5 | Channel compensated i-Vectors for DNN Speaker Recognition |
Albert Jiménez, TALP, Universitat Politècnica de Catalunya | |
P2-6 | UPC System in Mediaeval 2015 Multimodal Person Discovery in Broadcast TV |
Miquel Angel India Massana, TALP, Universitat Politècnica de Catalunya | |
P2-7 | Automatic Detection of Equipment Alarms in a Neonatal Intensive Care Unit Environment: a Knowledge-Based Approach |
Ganna Raboshchuk, TALP, Universitat Politècnica de Catalunya | |
P2-8 | A Deep Learning Approach to Automatic Detection of Equipment Alarms in a Neonatal Intensive Care Unit Environment |
Alex Peiró, TALP, Universitat Politècnica de Catalunya |
Latest news
Poster Award
Fernando de la Calle, from Universidad Carlos III de Madrid, has
been distinguished with the Best Student Poster Award for his work
Robustness of biologically motivated features for DNN-based
ASR.
Congratulations!
Poster Program Published
The Best Student Poster Award will be given on Thursday July 9, just before the Closing Ceremony +info
Venue
Due to the high demand, the lecture and lab rooms have been updated +info
Detailed Program Published
Registration to RTTHSS2015 is open
Register now - limited places