This workshop was held on 9 January 2016. You can download the proceedings via this link.
We had 3 keynote speakers:
Friday 8 January:
Conference dinner starting at 19:00 together with the participants of SPIN 2016.
Saturday 9 January:
09:00 - 09:45 Keynote:
Jonas Obleser: The listening brain and its adaptation to acoustic challenges
09:45 - 10:45 Oral presentations (1)
Odette Scharenborg, Elea Kolkman, Brechtje Post: Perception of Sentence Accent in Non-native Speech in Noise
Polina Drozdova, Roeland van Hout, Odette Scharenborg: Do noise and linguistic skills influence lexically-guided perceptual learning?
Jieun Song, Paul Iverson: The effect of accent on speech recognition in a competing-talker background: Neural oscillations and the N400
10:45 - 11:00 Break
11:00 - 12:00 Oral presentations (2)
Petra Hödl, Paul Iverson: Does acoustic unreliability of the speech signal affect N400?
Deepak Baby, Tuomas Virtanen, Hugo Van hamme: Coupled Dictionary-based Speech Enhancement with Adaptively Learned Atoms for the CHiME-3 Challenge
Johannes Zaar, Torsten Dau: Auditory features in consonant perception - a modeling perspective
12:00 - 13:00 Posters
13:00 - 14:00 Lunch
14:00 - 14:45 Keynote:
Deniz Baskent: Speech perception with cochlear implants: An inspiring challenge
14:45 - 15:45 Oral presentations (3)
Rebecca Carroll, Anna Warzybok: Speech recognition in reverberation and noise: The roles of age and the lexicon
Niels Henrik Pontoppidan, Marianna Vatti, Rikke Birksteen Rossing, Tom Barker, Tuomas Virtanen: Separating known competing voices for people with hearing loss
Hans Rutger Bosker, Eva Reinisch, Matthias Sjerps: Listening under cognitive load makes speech sound fast
15:45 - 16:15 Break / Posters (remain in the poster room)
16:15 - 17:15 Oral presentations (4):
Yue Zhang, Stuart Rosen: Listeners are still sensitive to the frequency distribution of cues in degraded speech, but only to the unimpaired ones
Kurt Steinmetzger, Stuart Rosen: The Role of Periodicity in Perceiving Speech in Quiet and in Background Noise with Simulated Cochlear Implants
Laurianne Cabrera, Lynne Werner: The use of slow and fast temporal envelope cues in phonetic discrimination at 3 months of age
17:15 - 18:00 Keynote:
Jonathan Simon: Neural Representations of Speech, and Speech in Noise, in Human Auditory Cortex
18:00: Close of workshop
Overview of poster presentations:
01. Emma Brint, Paul Iverson: The acoustic change complex in cochlear implant listeners
02. Huarda Valdes-Laribi, Dorothea Wendt, Ewen MacDonald, Martin Cooke, Sven Mattys: Sentence comprehension with a competing talker: informational interference or semantic facilitation?
03. Attila Máté Tóth, Martin Cooke, Jon Barker: Speech + noise = confusion: misperceptions arising from the interaction of speech and babble masker components.
04. Nemanja Cvijanovic Patrick Kechichian , Kees Janse , Armin Kohlrausch: On the use of ultrasound in speech communication systems
05. Dorina Strori, Johannes Zaar, Odette Scharenborg, Sven Mattys: Effects of Energetic Masking and Speech Modulation on Spoken Word Recognition
06. Juliane Schmidt, Odette Scharenborg, Diana Herzog, Esther Janse: Older adults' rating of affect-related prosody in speech: The roles of hearing loss and auditory processing abilities
07. Benjamin G. Schultz, Sonja Kotz: Finding the beat in German poetry: The role of meter, rhyme, and lexical content
08. Gusztáv Lőcsei, Sébastien Santurette, Torsten Dau, Ewen N. MacDonald: Binaural unmasking of speech with small interaural time differences and its relationship to binaural temporal fine structure processing
09. Florian Hintz, Odette E Scharenborg: Neighbourhood density influences word recognition in native and non-native speech recognition in noise
10. Ewa Wanat, Rachel Smith, Jane Stuart-Smith and Colin Hamilton: The role of rhythm in improving non-native comprehension of English casual speech
11. Sara Ahmadi, Bert Cranen, Lou Boves: Guiding ASR towards HSR performance using modulation spectrum features.
12. Gustav Henter: Measuring the perceptual effect of speech synthesis modelling assumptions
13. Ricard Marxer: Data and Tools for Microscopic Intelligibility Modelling
Humans adapt well to sensory degradation. In order to do so, our cognitive strategies need to adjust accordingly. The auditory sensory modality poses an excellent, although under-utilised, research model to understand these adjustments, their neural bases, and their large variation amongst individuals.
In my talk I will present recent findings from our laboratory, utilizing mainly the power and phase of neural oscillations as measured in magneto- and electroencephalography (M(EEG). I will present initial evidence for a useful distinction of broadly “automatic” or effortless neural mechanisms (e.g., neural entrainment) that allow for improved performance in challenging listening situations versus broadly “controlled” or effortful neural mechanisms (e.g., alpha power enhancement in domain-general networks). I will argue that both mechanisms support listening performance, and I will link our results to ageing and hearing loss.
Cochlear implants are prosthetic hearing devices that help deaf individuals regain hearing. The device is a wonderful example of a successful neural prosthesis, through which we also learn more about perception of speech in normal hearing. Yet, there still remain a number of unsolved problems, particularly the difficulty in perception of speech via an implant in complicated listening environments, as well as the high variation in the outcome from implantation. One of the main challenges is that the list of contributing factors is relatively large (device-related factors, physiology/etiology, cognitive factors, aging, just to name a few). Additionally, the commonly used clinical tests for word and sentence identification seem to only partially reveal the underlying mechanisms of speech perception with implants. In this talk, I will present our recent work where we employ a systematic and multidisciplinary approach to move beyond clinical measures of intelligibility, and to better capture speech perception mechanisms with a cochlear implant in realistic listening situations.
We investigate how continuous speech, whether presented alone, degraded with noise, or masked by other speech signals, is represented in human auditory cortex. We use magnetoencephalography (MEG) to record the neural responses of listeners to continuous speech in a variety of contexts. We find that cortical representations of continuous speech are robust to noise under a wide variety of conditions, including clean speech, additive stationary noise, band-vocoded speech and noise, and speech with competing speakers. In the last case, individual neural representations of the speech of both the foreground and background speaker are observed, with each being selectively phase locked to the rhythm of the corresponding speech stream. In all observed cases, the temporal envelope of the acoustic speech stream can be reconstructed from the observed neural response to the speech.