Table 7. Summary performance of the S-AMPH model.

The table shows accuracy of performance for acoustic-phonological mapping at each phonological level for timed and untimed child-directed speech, and the advantage for timed-untimed CDS. The mean and standard deviation of the distance of the models' estimates from their respective phonological targets is also included.

		Timed CDS	Untimed CDS	Timed-Untimed
Syllable	Accuracy	98.3%	81.8%	16.5%
	d'	4.33	1.82	2.51
	Distance from target (SD)	22.8 ms (21.4 ms)	7.52 ms (30.7 ms)	15.28 ms
Onset-Rime	Accuracy	90.9%	77.6%	13.3%
	d'	2.76	1.53	1.23
	Distance from target (SD)	-1.80 ms (10.9 ms)	1.45 ms (11.2 ms)	-3.33 ms
Prosodic Stress	Accuracy	94.8%	72.4%	22.4%
	d'	3.46	1.21	2.25