Speech Adaptation in Extended Ambient Intelligence Environments
Authors: Bonnie Dorr, Lucian Galescu, Ian Perera, Kristy Hollingshead-Seitz, David Atkinson, Micah Clark, William Clancey, Yorick Wilks, Eric Fosler-Lussier
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Initial data collection of impaired speech for testing will consist of de-identified speech recordings gathered from quarterly visits over 3 years of 25 ALS patients currently followed by James A. Haley Veteran’s Administration Hospital (Tampa VA). We will record a mixture of context-independent calibration sentences and context-dependent conversational speech regarding events in the patients lives. These controlled data will be essential in quantifying the various speech and biological factors that change over time so that we have a base to build upon and compare performance to when generalizing our models to an uncontrolled ambient environment |
| Researcher Affiliation | Academia | Institute for Human and Machine Cognition, 15 SE Osceola Avenue, Ocala, FL {bdorr, lgalescu, iperera, khollingshead, datkinson, mclark, wclancey, ywilks}@ihmc.us Eric Fosler-Lussier The Ohio State University, 2015 Neil Avenue, Columbus, OH fosler@cse.ohio-state.edu |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | No | The paper describes initial data collection for testing, stating: "Initial data collection of impaired speech for testing will consist of de-identified speech recordings gathered from quarterly visits over 3 years of 25 ALS patients currently followed by James A. Haley Veteran’s Administration Hospital (Tampa VA)." This is a newly collected, internal dataset, with no information provided for public access. |
| Dataset Splits | No | The paper mentions data collection for testing but does not specify any training, validation, or test dataset splits (e.g., percentages or sample counts). |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU models, CPU types, or memory specifications used for running experiments. |
| Software Dependencies | No | The paper states: "We anticipate using Kaldi (Povey et al. 2011) as our base speech recognition system", but it does not specify a version number for Kaldi or any other software dependencies. |
| Experiment Setup | No | The paper describes an "Experimental Design" section that focuses on data collection and environmental monitoring, but it does not provide specific details about hyperparameter values, training configurations, or other system-level settings for any models or experiments. |