Satellite Workshops

The five satellite workshops listed below are offered as ancillary meetings for those attending InterSpeech 2012 or anyone who is interested.  Satellite Workshops are not included in the conference registration fee. To attend, you must register with the Satellite Workshop directly and make all necessary arrangements per the instructions on their respective websites.

Interdisciplinary Workshop on Feedback Behaviors in Dialog

Friday, September 7 – Saturday, September 8, 2012

Website: http://www.cs.utep.edu/feedback

Location: Skamania lodge, Columbia River Gorge

1131 SW Skamania Lodge Way

Feedback skills are important for people (or machines) wishing to be able to function as supportive, cooperative listeners. The production and comprehension of back-channels and related phenomena, including response tokens, reactive tokens, minimal responses, continuers and acknowledgments, are also of scientific interest, as possibly the most accessible example of the real-time responsiveness that underpins many successful interpersonal interactions. This workshop will provide a venue for an interdisciplinary examination of these phenomena and feedback behaviors in dialog.

SAPA-SCALE

Friday, September 7 – Saturday, September 8, 2012

Website: http://www.sapaworkshops.org/2012

Location: Hilton, Portland, Oregon

921 SW Sixth Avenue

Papers are solicited for the SAPA-SCALE conference, an ISCA-supported event to be held as a satellite to Interspeech 2012 in Portland, USA, September 2012. Following the successful Workshops on Statistical and
Perceptual Audition (SAPA 2004, 06, 08, 10), the event this year is organized jointly with the Speech Communication with Adaptive Learning (SCALE) consortium. The principal objective of the conference is to bring together researchers addressing perceptually motivated speech and audio processing tasks with the tools of statistical signal processing and machine learning. The themes of the conference are:

• Statistical models for speech and audio processing motivated by human perception
• Developing the commonalities between speech recognition and synthesis to provide richer and more sophisticated models for speech
• Adaptive learning approaches to speech and audio signal processing and their incorporation into statistical models

This will be a two-day, single-track conference with an informal atmosphere structured to promote discussion. There will be keynotes from leading researchers in addition to a limited number of oral presentations chosen for breadth and provocation. All participants will be actively engaged in the effort to broaden perspectives and foster novel research directions and interesting variants on current approaches.

Blizzard Challenge

Friday, September 14, 2012

Website: http://www.synsig.org/index.php/Blizzard_Challenge_2012

Location: Hilton, Portland, Oregon

921 SW Sixth Avenue

In order to better understand and compare research techniques in building corpus-based speech synthesizers on the same data, the Blizzard Challenge 2012 will be held. The basic challenge is to take the released speech data, build synthetic voices, and synthesize a prescribed set of test sentences. The sentences from each synthesizer are then evaluated through extensive listening tests. The results will be presented at Blizzard Challenge Workshop 2012 in Portland.

WOCCI - 3rd Workshop on Child, Computer and Interaction

Friday, September 14, 2012

Website: http://www.wocci.org

Location: Vey Auditorium, OHSU Marquam Hill Campus, Portland, Oregon

The WOCCI Workshop, the third one so far, aims to bring together researchers and practitioners from universities and industry working in all aspects of child-machine interaction including computer, robotics and multi-modal interfaces. Human-machine interaction poses significantly unique challenges and problems in the case of children, both in modeling the acoustic/linguistic components as well as the interactivity.

The Workshop will provide an opportunity for researchers from disparate communities such as cognitive science, robotics, speech processing, linguistics and application areas to discuss common solutions and frameworks child-machine interaction. These may include current state-of-the-art components as well as specifications of next generation interface. We expect the workshop will help address societal needs such as the delivery of health and education through computer interactions. For example, to deliver personalized remedial treatments for children with or without disabilities. The Workshop will encompass recent advancements in core technologies as well as experimental systems and prototypes.

MLSLP2012 - The Second Symposium on Machine Learning in Speech and Language Processing

Friday, September 14, 2012

Website: http://www.ttic.edu/sigml/symposium2012/

Location: Hilton, Portland, Oregon

921 SW Sixth Avenue

The goal of the symposium is to foster communication and collaboration between researchers in machine learning and speech and language processing. It is the second annual meeting of the ISCA Machine Learning Special Interest Group (SIGML).  The workshop will feature both invited talks and general submissions. We solicit a broad array of submissions, including novel research and, especially, position and review papers addressing topics that are relevant to the speech, machine learning, and natural language processing research communities. These areas include, but are not limited to, applications to speech/NLP of log-linear models, support vector machines, neural networks, kernel methods, discriminative transforms, large-margin training, discriminative training, active/semi-supervised/unsupervised learning, structured prediction, Bayesian modeling, deep learning, and sparse representations.