Speech acoustics

Speech Acoustics. With the advent of the computer age, speech synthesis moved to digital processing, either by bandwidth compression of stored speech or by using a speech synthesizer. The synthesizer reads a text and then produces the appropriate phonemes on demand from their basic acoustic parameters, such as the vibration frequency of the vocal cords and the frequencies and amplitudes of the vowel formants. This method of generating speech is considerably more efficient in terms of data storage than archiving a dictionary of prerecorded phrases. 

Another important, and probably the most difficult, area of speech acoustics is the machine recognition... 

G. Demoment, IEEE Trans, on Acoust., Speech, Signal Processing 37 12, pp. 2024-2036 (1989)... 

Reverberation Control. Reverberation time (T q) is defined as the length of time in seconds for the sound of an instantaneously stopped source in a room to decay by 60 decibels (dB). Reverberation time is one important factor in determining the acoustical character of a space and its suitabiHty for specific activities. For lectures and other speech activities a relatively short reverberation time is desirable so that syllables do not persist and overlap one another, causing difficulty with inteUigibiHty conversely, for music activities, a relatively long reverberation time is desirable to allow blending of the sound and a sense of being surrounded by the music. Without reverberation music usually sounds dull and lifeless. 

R. F. Lyon and E. P. Loeb, Isolated Digit Recognition Experiments with a Cochlear Model, in Proceedings International Conference on Acoustics Speech and Signal Processing—ICASSP-87, Dallas, Texas (April 1987). 

T. Martin, Acoustic Recognition of Limited Vocabulary in Continuous Speech, Dept. Electrical Engineering, Univ, Pennsylvania, Ph.D. Thesis (1970). 

Gallagher, N. C., Jr., and Wise, G. L., A theoretical analysis of the properties of median filters, IEEE Trans. Acoustics, Speech, Signal Proc. 29, 1136 (1981). 

Waibel, A., Hanazawa, T., Hinton, G., Shikano, K. Lang K..J. 1989. Phoneme recognition using time-delay neural networks. IEEE Transactions On Acoustics, Speech, and Signal Processing, 37/3, 328-339. 

Hunt, B. R., and Trussell, H. J. (1976). Proc. Int l. Conf. Acoustics, Speech, and Signal Processing, Philadelphia, April, IEEE. pp. 354-356. 

Thomas, G. (1981). Proc. IEEE Conf. on Acoustics, Speech and Signal Processing, Vol. 1, pp. 47-49. 

Shore, J. E. (1981). IEEE Trans. Acoust. Speech, Sig. Process. ASSP-29, 230-237. 

The excitation signal depends on the source model (vocal cord models are used in speech for both voice and unvoiced sources). LPC models can be used in other circumstances, for example, the modeling of acoustic musical instruments where the... 

Almeida and Silva, 1984b] Almeida, L. and Silva, F. (1984b). Variable-frequency synthesis An improved harmonic coding scheme. In Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing., San Diego, CA. 

Bauer and Seitzer, 1989b] Bauer, D. and Seitzer, D. (1989b). Statistical properties of high quality stereo signals in the time domain. In Proc. IEEE Int. Conf. Acoust., Speech and Signal Proc, pages 2045 - 2048. 

Clark et al., 1981b] Clark, G., Mitra, S., and Parker, S. (1981b). Block implementations of adaptive digital filters. IEEE Trans. Acoust. Speech and Sig. Proc., ASSP-29 744-752. 

---