Low-cost Speech Recognition Research Articles

Compact Speech Recognition Algorithm Based on Vowel Identification and Its Evaluation on a DSP Board

This paper describes a low-cost speech recognition system for controlling terminals and systems. The speech recognition system uses a compact speech recognition algorithm that is based on symbol matching for the five Japanese vowel sounds and a few consonant categories instead of conventional speech parameters. The algorithm has a significant advantage over conventional speaker-independent speech recognition algorithms that use complicated matching techniques; it reduces the memory required (by a factor from 4 to 200) for storing the reference templates and instruction sets and so can be implemented on low-cost processors. We implement this recognition algorithm on a fixed-point, 20-MIPS digital signal processor board with 10-k x 16-bit on-chip memory. Recognition experiments using 10 (20) Japanese city names indicate an accuracy of 96.2% (90.3%). By restricting the target vocabulary 20 words, the proposed speech recognition system can be integrated into many devices at Very low cost.

Low-cost speech recognition system and method

A low cost speech recognition system generates frames of received speech having binary feature components. The received speech frames are compared (18) with reference templates (22) , and error values representing the difference between the received speech and the reference templates (22) are generated. At the end of an utterance, if one template resulted in a sufficiently small error value, the word represented by that template is selected (26) as the recognized word.

Statistical clustering procedures applied to low-cost speech recognition

A new generation of low-cost speech recognition devices are appearing, which offer much promise for useful applications in biomedical engineering. These devices are statistical pattern recognizers. Input utterances are classified by comparison with a set of templates derived during ‘speaker training’. For useful application of these devices, recognition accuracy must be high and speaker training must not be unacceptably complicated or tedious. This paper investigates techniques which consider the statistical nature of the input utterances, used to improve recognition accuracy. Word classification based on the Mahalonobis distance metric, and using templates derived from cluster analysis of the training inputs, was found to give results superior to the other strategies studied. This classifier was unsuitable for implementation in a real-time, low-cost system but the principle of clustering was successfully applied to produce an adaptive system which tracked changes in the user's voice. This allowed training to be drastically simplified by updating templates during normal operation. The adaptive system achieved 98.8% recognition accuracy on a 32 word vocabulary compared to 94.8% without adaptation.