Word recognition, independent of the speaker, without segmentation
In automatic word-recognition, experience has shown the difficulties of making the correct segmentation. Incorrect segmentation often leads to false recognition. The segmentation problem can be avoided if use is made of the short-time spectrogram (“Sonagram”, “Voice-print”) of a complete word for recognition, together with operating a method of word-building from the voice-print. In the present work the two-dimensional Fourier coefficients (“Room-frequencies”) of the spectrograms are considered. The Fourier coefficients with low indices here have the considerable advantage of being relatively independent of the rapidity of speech, of the fundamental frequency and of the precise location of the formant frequency. From information gained from many speakers the average characteristic parameter-vector for each word is obtained. Newly acquired words are used for the recognition of the same speakers. The parameter vectors of these are compared with each of the sample vectors whereby three different separation measurements are applied. For the best of the three for recognition separation measurements were different words (the number-words of the German language) from altogether 600 phrases 589 (over 98%) were recognised correctly.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: November 1, 1976
More about this publication?
- Acta Acustica united with Acustica, published together with the European Acoustics Association (EAA), is an international, peer-reviewed journal on acoustics. It publishes original articles on all subjects in the field of acoustics, such as general linear acoustics, nonlinear acoustics, macrosonics, flow acoustics, atmospheric sound, underwater sound, ultrasonics, physical acoustics, structural acoustics, noise control, active control, environmental noise, building acoustics, room acoustics, acoustic materials, acoustic signal processing, computational and numerical acoustics, hearing, audiology and psychoacoustics, speech, musical acoustics, electroacoustics, auditory quality of systems. It reports on original scientific research in acoustics and on engineering applications. The journal considers scientific papers, technical and applied papers, book reviews, short communications, doctoral thesis abstracts, etc. In irregular intervals also special issues and review articles are published.
- Editorial Board
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Information for Advertisers
- Online User License
- Ingenta Connect is not responsible for the content or availability of external websites