The Human Speechome Project (by analogy to "genome" and similar terms) is an effort to closely observe and model the language acquisition of a child over the first three years of life.
The project was conducted at the Massachusetts Institute of Technology's Media Laboratory by the Associate Professor Deb Roy[1] with an array of technology that is used to comprehensively but unobtrusively observe a single child – Roy's own son[2] – with the resulting data being used to create computational models to yield further insight into language acquisition.
[3] Most studies of human speech acquisition in children have been done in laboratory settings and with sampling rates of only a couple of hours per week.
Frank Moss, director of the Media Lab[4]A digital network consisting of eleven video cameras, fourteen microphones, and an array of data capture hardware was installed in the home of the subject.
[5] Data was gathered at an average rate of 200 gigabytes per day, necessitating the development of sophisticated data-mining tools to reduce analysis efforts to a manageable level, and transcribing significant speech added a labor-intensive dimension.