1.   Areas such as vision, continuous speech recognition and synthesis, and machine learning have been hard.

2.   Continuous speech recognition and synthesis are additional examples of tasks neural networks are undertaking with reasonable success.

3.   Moreover, when words are spoken in continuous speech they often sound different from when spoken in isolation.

4.   Possible applications are continuous speech recognition and commands to robot arms.

5.   Continuous speech processing requires the application of very many sources of knowledge in order to decode the utterance.

6.   These experiments show that the recognition strategies suggested by structural analysis of large lexicons do not, in their present form, extend to continuous speech recognition.

7.   One of the aims of this chapter is to look at the effect on word discriminability in continuous speech of increasing the depth of the phoneme graph.

8.   However, when we look at continuous speech in English utterances we find that these tones can only be identified on a small number of particularly prominent syllables.

9.   All these programs take dictation in continuous speech.

10.   Both recently have introduced sophisticated systems for medical transcription that understand continuous speech.