The two tables below illustrate the different ways the pitch accents, phrase accents, and boundary tones of American English combine in phrase-final position. The different contours are exemplified in the top table with the four-syllable proper noun Menomonee and in the bottom table with the monosyllabic interjection no. Both words are composed of entirely of sonorants (i.e. vowels and nasal consonants), thus minimizing consonantal perturbations in the F0 track.

The representation here follows the original (circa 1980s) version of the system. In this system, the difference between H*+L and H* is neutralized before a low phrase tone (i.e. in L-L% and L-H%), hence the two 'N/A' cells in each table. For an updated version (based on the revision to the system in the 1990s), see this page.

Images:

Each image contains an F0 track superimposed over a spectrogram, both corresponding to the same soundfile. Both the F0 track and the spectrogram share the same time domain, zoomed into the actual linguistic contents of the soundfile (hence the x axis does not always begin at 0). The y axis indicates F0 (in Hertz) and has been fine-tuned to facilitate comparison within each row. (To see the axis labels more clearly, you may want to zoom in slightly.)

The size of each dot making up the F0 track correspond to the intensity at that time point (with larger dots indicating greater intensity), thereby highlighting the location of stress in each utterance. The spectrograms were generated with a dynamic range of 40 dB and a frequency range (y axis) of 0-5500 Hz.

Audio:

If you are using the most up-to-date version of your Internet browser, an audio player will display beneath each image, which you can click to play the corresponding soundfile. Any browser will work, although the large size of the audio player in Internet Explorer may make things hard to see.