Estonian Words Research Articles

Scores for connected recognition of words differing in distinctive quantity

The results of an experiment on talker-dependent, connected-speech recognition of 10 Estonian words are reported. The words differ in so-called distinctive quantity, the major acoustic correlate of which is duration. These were consonant-vowel-vowel words, each word modeled with four variable-duration states. The words were spoken, and recognized, in sentence pairs of the form 'Did you say (word 1, word 2, word 3)? No, I said (word 4, word 5, word 6)'. The test sentences were spoken either at the same rate as the training sentences or at a much faster rate. In a first set of experiments, the likelihood of the spectral match was the only type of factor in the dynamic-programming best-path score. The best word recognition results obtained were 8% and 64% on the slow and fast test sentences respectively. In a second set of experiments, on probabilities or likelihoods of state durations, the best word recognition results were 86% and 68% respectively. It is concluded that speech rate can be a major problem for connected recognition of these words, and that in these experiments the problem has not been completely overcome, even using the likelihoods of the state duration ratios.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The Function of Quantity in Finnish and Estonian

The purpose of this paper1 is to compare some aspects of the quantity systems of two related languages, Finnish and Estonian. Finnish and Estonian belong to the Balto-Finnic branch of the Finno-Ugric language family. They have almost identical segmental phoneme inventories. Superficially it also seems that they have very similar word structures. Often Finnish and Estonian words can be matched morpheme for morpheme or even phoneme for phoneme. Both languages have fixed primary stress on the first syllable of a word.2 In principle, both are quantity languages: oppositions between short and long phonemes permeate the phonological structure of both. Upon closer scrutiny, however, the quantity systems of the two languages show farreaching differences. Finnish has two degrees of quantity.3 Every vowel is either long or short. Diphthongs count as equivalent to long vowels. Short and long vowels contrast in any syllable of a word; there are no positional restrictions, either with regard to place within the word or with regard to stress. The opposition between short and long consonants can be realized only in intervocalic position; there are no consonantal quantity contrasts in word-initial and word-final position. Long consonants in intervocalic position contain a syllable boundary and are distributed between the two syllables so that the first part of the consonant closes the preceding syllable and the second part starts the following syllable. Consonant clusters function, in general, in a similar manner.4 No clusters occur in wordinitial or word-final position, the positions where consonant quantity is noncontrastive. There are numerous segmental restrictions which eliminate some segmental sequences and influence the relative frequency of other combinations. The

On quantity and stress in Estonian

Abstract According to the traditional theory, the distinctive opposition which occurs in the Estonian words sada “hundred”-saada “send, impt.”- `saada “to get, receive, inf.”, lina “flax”-g. linna-p. `linna “town”, g. silma-p. ` silma “eye” represents the different grades of quantity or duration of phonemes /a/, /n/ and /l/, i. e. in Estonian the phonemes have three degrees of quantity or length: the first, second and third degree (henceforth denoted as I, II, III). This theory was for the first time put forward by O. W.Masing in 1824. 1 Besides the length of single phonemes one also speaks of the length of phoneme clusters: in the word forms g. seina-p. `seina “wall” the diphthong in the first form is medium long or in II, in the second form extra long or in III; in the forms g. silma-p. ` silma the consonant cluster /lm/ in the first form is in II, in the second form in III.