Readings in Speech Recognition

After more than two decades of research activity, speech recognition has begun to live up to its promise as a practical technology and interest in the field is growing dramatically. Readings in Speech Recognition provides a collection of seminal papers that have influenced or redirected the field and that illustrate the central insights that have emerged over the years.

The editors provide an introduction to the field, its concerns and research problems. Subsequent chapters are devoted to the main schools of thought and design philosophies that have motivated different approaches to speech recognition system design. Each chapter includes an introduction to the papers that highlights the major insights or needs that have motivated an approach to a problem and describes the commonalities and differences of that approach to others in the book.

In 1990, most speech recognition was of single words, not continuous speech, and it was of a given speaker. That is, it was not speaker independent. Plus, due to the limited memory and slow cpus, often the analysis was not in anything approaching realtime. Typically, the speaker would say something, word by word, and this would be recorded in digital form, which would then be analysed.

Even with these hardware limitations, the papers describe promising approaches and indeed of good progress in the subject. Which is actually what did happen subsequently.

As an aside, Kai-Fu Lee came to prominence at Carnegie Mellon in the late 80s, writing key parts of the Sphinx speech recognition system, which was highly regarded as the benchmark of its time.