A Statistical Language Modeling Approach to Lattice-Based Spoken Document Retrieval

Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 810–818, Prague, June 2007. c⃝2007 Association for Computational Linguistics A Statistical Language Modeling Approach to Lattice-Based Spoken Document Retrieval Tee Kiah Chia† Haizhou Li‡ Hwee Tou Ng† †Department of Computer Science National University of Singapore 3 Science Drive 2, Singapore 117543 {chiateek,nght}@comp.nus.edu.sg ‡Institute for Infocomm Research 21 Heng Mui Keng Terrace Singapore 119613 hli@i2r.a-star.edu.sg Abstract Speech recognition transcripts are far from perfect; they are not of sufficient quality to be useful on their own for spoken document retrieval. This is especially the case for con- versational speech. Recent efforts have tried to overcome this issue by using statistics from speech lattices instead of only the 1- best transcripts; however, these efforts have invariably used the classical vector space re- trieval model. This paper presents a novel approach to lattice-based spoken document retrieval using statistical language models: a statistical model is estimated for each doc- ument, and probabilities derived from the document models are directly used to mea- sure relevance. Experimental results show that the lattice-based language modeling method outperforms both the language mod- eling retrieval method using only the 1-best transcripts, as well as a recently proposed lattice-based vector space retrieval method. 1 Introduction Information retrieval (IR) is the task of ranking a collection of documents according to an estimate of their relevance to a query. With the recent growth in the amount of speech recordings in the form of voice mails, news broadcasts, and so forth, the task of spoken document retrieval (SDR) – information retrieval in which the document collection is in the form of speech recordings – is becoming increas- ingly important. SDR on broadcast news corpora has been “deemed to be a solved problem”, due to the fact that the performance of retrieval engines working on 1- best automatic speech recognition (ASR) transcripts was found to be “virtually the same as their perfor- mance on the human reference transcripts” (NIST, 2000). However, this is still not the case for SDR on data which are more challenging, such as conver- sational speech in noisy environments, as the 1-best transcripts of these data contain too many recogni- tion errors to be useful for retrieval. One way to ameliorate this problem is to work with not just one ASR hypothesis for each utterance, but multiple hy- potheses presented in a lattice data structure. A lat- tice is a connected directed acyclic graph in which each edge is labeled with a term hypothesis and a likelihood value (James, 1995); each path through a lattice gives a hypothesis of the sequence of terms spoken in the utterance. Each lattice can be viewed as a statistical model of the possible transcripts of an utterance (given the speech recognizer’s state of knowledge); thus, an IR model based on statistical inference will seem to be a more natural and more principled approach to lattice-based SDR. This paper thus proposes a lattice-based SDR method based on …

Content truncated to 500 words