Name
Title | ||
|---|---|---|
SpeechFind: Advances in Spoken Document Retrieval for a National Gallery of the Spoken Word |
Abstract | ||
|---|---|---|
Advances in formulating spoken document retrieval for a new National Gallery of the Spoken Word (NGSW) are addressed. NGSW is the first large-scale repository of its kind, consisting of speeches, news broadcasts, and recordings from the 20th century. After presenting an overview of the audio stream content of the NGSW, with sample audio files from U.S. Presidents from 1893 to the present, an overall system diagram is proposed with a discussion of critical tasks associated with effective audio information retrieval. These include advanced audio segmentation, speech recognition model adaptation for acoustic background noise and speaker variability, and information retrieval using natural language processing for text query requests that include document and query expansion. For segmentation, a new evaluation criterion entitled fused error score (FES) is proposed, followed by application of the CompSeg segmentation scheme on DARPA Hub4 Broadcast News (30.5% relative improvement in FES) and NGSW data. Transcript generation is demonstrated for a six-decade portion of the NGSW corpus. Novel model adaptation using structure maximum likelihood eigenspace mapping shows a relative 21.7% improvement. Issues regarding copyright assessment and metadata construction are also addressed for the purposes of a sustainable audio collection of this magnitude. Advanced parameter-embedded watermarking is proposed with evaluations showing robustness to correlated noise attacks. Our experimental online system entitled “SpeechFind” is presented, which allows for audio retrieval from a portion of the NGSW corpus. Finally, a number of research challenges such as language modeling and lexicon for changing time periods, speaker trait and identification tracking, as well as new directions, are discussed in order to address the overall task of robust phrase searching in unrestricted audio corpora. |
Year | DOI | Venue |
|---|---|---|
2005 | 10.1109/TSA.2005.852088 | Speech and Audio Processing, IEEE Transactions |
Keywords | Field | DocType |
audio signal processing,information retrieval,natural languages,speech recognition,watermarking,CompSeg segmentation scheme,DARPA Hub4 Broadcast News,National Gallery of the Spoken Word,SpeechFind,acoustic background noise,advanced audio segmentation,advanced parameter-embedded watermarking,audio information retrieval,copyright assessment,correlated noise attacks,fused error score,language modeling,metadata construction,natural language processing,overall system diagram,robust phrase searching,speaker variability,speech recognition model,spoken document retrieval,structure maximum likelihood eigenspace mapping,sustainable audio collection,text query requests,transcript generation,unrestricted audio corpora,Accent classification,broadcast news,document expansion,environmental sniffing,fidelity,fused error score,information retrieval,language modeling,model adaptation,query expansion,robust speech recognition,robustness,security,speech segmentation,spoken document retrieval,watermarking | Query expansion,Phrase search,Audio mining,Computer science,Speech recognition,Artificial intelligence,Natural language processing,Document retrieval,Speech segmentation,Audio signal processing,Language model,Cable television | Journal |
Volume | Issue | ISSN |
13 | 5 | 1063-6676 |
Citations | PageRank | References |
39 | 1.75 | 74 |
Authors | ||
8 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| John H. L. Hansen | 1 | 3215 | 365.75 |
| Rongqing Huang | 2 | 141 | 10.27 |
| Bowen Zhou | 3 | 2212 | 246.21 |
| Michael S. Seadle | 4 | 72 | 12.52 |
| J. R. Deller | 5 | 104 | 9.16 |
| Aparna Gurijala | 6 | 55 | 6.21 |
| Mikko Kurimo | 7 | 908 | 93.37 |
| Pongtep Angkititrakul | 8 | 179 | 15.47 |