Writing and Speech Recognition
Observing Error Correction Strategies of Professional Writers of Professional Writers
How does speech recognition influence the text production process of novice and experienced speech recognition users? How do professional writers deal with the special characteristics of speech recognition? How can multimodal writing processes be observed and analyzed? This book examines these questions, and others.
Compared to speaking, writing is far more complex. Writing can be seen as a visual representation of spoken language: a combination that speech recognition takes full advantage of. However, speech recognition causes unfamiliar errors in the text produced so far. Consequently, writers develop various strategies to solve these errors. Also, different error types seem to lead to different error correction strategies.
Research in speech recognition has a double agenda: it examines the organization of speech recognition based writing processes speech recognition as a research object but also employs speech recognition as a research instrument. This double approach is demonstrated in the various research projects in this thesis.