RECOGNITION OF PHONEMES AND WORDS IN SINGING

This paper studies the influence of n-gram language models in the recognition of sung phonemes and words. We train uni-, bi-, and tri-gram language models for phonemes and bi- and trigrams for words. The word-level language model is estimated from a textual lyrics database. In the recognition we use a hidden Markov model based phonetic recognizer adapted to singing voice. The models were tested on monophonic singing and on vocal lines separated from polyphonic music. On clean singing the phoneme recognition ac-curacies varied from 20 % (no language model) to 39 % (bigram) and on polyphonic music from 6 % (no language model) to 20 % (bi-gram). In word recognition, one fifth of the words were recognized in clean singing, the performance being lower on polyphonic music. We study the use of the recognition results in a query-by-singing application. Using the recognized words, we retrieve the songs by searching for the text in a text lyrics database. For the word recog-nition system having only 24 % correct recognition rate, the first re-trieved song is correct in 57 % of the test cases. Index Terms — singing recognition, speech recognition, query-by-singing 1