Speech and Hearing Sciences ETDs

Publication Date



The validity of The Oral Nasal Acoustic Ratio (Tonar) System as a means of obtaining data relevant to perceived nasality was established by comparing forward and backward play scale values with nasal/oral acoustic ratios computed by Tonar.

The acoustic characteristics of orally and nasally emitted signals and their relative contributions to the total speech signal were analyzed and considered within the framework of oral and nasal acoustic masking.

A passage free of nasal phonemes was read by 20 subjects with surgically repaired palatal clefts. A specially designed instrumentation system (Tonar) was used to separate the signals emerging from the oral and nasal tracts for recording on a two-channel tape recorder. The separate identity of the signals was maintained as they were amplified and analyzed by interlocked frequency and intensity analyzers set to cover frequencies from 250 to 2750. The signals thus analyzed were channeled into an analog ratio computer where the voltage levels representing the nasal and oral signals were compared and the acoustic ratio computed. The contin­uously computed ratios were then displayed on oscillographic paper from which digital data were derived. To analyze the acoustic contributions of orally and nasally emitted signals, 6 of the 21 recorded speech samples were selected for further analysis. A sample phrase, arbitrarily chosen from the passage, was subjected to detailed frequency analysis by Tonar. To gain a general impression of relative nasality, a 2000 Hz frequency band covering frequencies from 250 Hz to 2250 Hz was used to measure nasal and oral intensities and to compute ratios for each of the six subjects. A 200 Hz bandwidth was used to examine the Tonar ratio as a function of frequency. All frequencies between 150 Hz and 3250 Hz of the same sample were analyzed and the nasal-to­oral acoustic ratios for each of the 16 pass bands were computed.

The data revealed: (1) Narrowband nasal/oral ratios varied as a function of frequency and appeared to be the result of a variety of interactions between oral and nasal intensities. (2} Oral and nasal energy concentrations were centered around a frequency of 500 Hz. (3) Narrowband acoustic ratios for central frequencies of 350, 550, and 750 were the principal contributors to the wideband ratio.

It was concluded that the potential for oral or nasal masking was present. It was further suggested that perceived nasality may be the result of variable nasal masking of the underlying oral signal. Conversely, nasality may be present but undetected as a result of oral masking.

Should nasal masking of the oral signal be a reality, it could mean that nasality, in addition to being a distrac­tive quality, may be covering (masking) acoustic cues vital to phonemic discriminations.

Degree Name

Speech-Language Pathology

Level of Degree


Department Name

Speech and Hearing Sciences

First Committee Member (Chair)

Samuel Glen Fletcher

Second Committee Member

Lloyd Edmond Lamb

Third Committee Member

Richard Baxter Hood

Document Type