Factors affecting speech retrieval

Collections of speech documents can be searched using speech retrieval, in which the documents are processed by a speech recogniser to give text that can be searched by standard text retrieval techniques. Recognition is the translation of speech signals into either words or subword units such as phonemes. We investigated the use of a phoneme-based recogniser to obtain phoneme sequences. We found that phoneme recognition is worse than word recognition, because of lack of context and di culty in phoneme boundary detection. Comparing the transcriptions of two di erent phonemebased recogniser, we found that the e ects of training using well-de ned phoneme data, the lack of a language model, and lack of a context-dependent model a ected recognition performance. Retrieval was based on n-grams. We found that trigrams performed better than quadgrams because the longer n-gram features contained too many transcription errors. Comparing the phonetic transcriptions from a word recogniser to transcriptions from a phoneme recogniser, we found that using 61 phones modelled with an algorithmic approach were better than using 40 phones modelled with a dictionary approach.