Phonetic searching vs. LVCSR: How to find what you really want in audio archives

International Journal of Speech Technology

Volumn 5, Issue 1, 2002, Pages 9-22

  Cardillo, Peter S ^a   Clements, Mark ^a   Miller, Michael S ^a  

^a Fast Talk Communications Inc   (United States)

Author keywords

Digital media asset management (DMAM); Large vocabulary continuous speech recognition (LVCSR); Phonetic searching

Indexed keywords

METADATA; ONLINE SEARCHING; SPEECH ANALYSIS; SPEECH COMMUNICATION;

DIGITAL MEDIA ASSET MANAGEMENT (DMAM); PHONETIC SEARCHING;

CONTINUOUS SPEECH RECOGNITION;

EID: 0036171616     PISSN: 13812416     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1013670312989     Document Type: Article

References (18)

1. Improving wordspotting performance with artificially generated data
2. SCAN - Speech content based audio navigator: A systems overview
3. Phonetic searching of digital audio
4. The TREC spoken document retrieval track: A success story
5. The 1996 broadcast news speech and language-model corpus
6. Microsoft windows highly intelligent speech recognizer: WHISPER
7. A fast lattice-based approach to vocabulary independent wordspotting
8. Spoken document retrieval for TREC-8 at Cambridge University
9. Phonetic recognition for spoken document retrieval
10. Phonetic set indexing for fast lexical access
11. Automatic recognition of keywords in unconstrained speech using Hidden Markov Models
12. The enhancement of wordspotting techniques