Title :
Statistical language model adaptation for Mandarin broadcast news transcription
Author :
Chen, Berlin ; Tsai, Wen-Hung ; Kuo, Jen-Wei
Author_Institution :
Graduate Inst. of Comput. Sci. & Inf. Eng., Nat. Taiwan Normal Univ., Taipei, Taiwan
Abstract :
This paper investigates statistical language model adaptation for Mandarin broadcast news transcription. A topical mixture model was proposed to explore the long-span latent topical information for dynamic language model adaptation. The underlying characteristics and various kinds of model complexities were extensively investigated, while their performance was verified by comparison with the conventional MAP-based adaptation approaches, which are devoted to extracting the short-span n-gram information. Speech recognition experiments were conducted on the broadcast news collected in Taiwan. Very promising results in both perplexity and word error rate reductions were initially obtained.
Keywords :
natural languages; speech recognition; statistical analysis; Mandarin broadcast news transcription; TMM-based method; dynamic language model adaptation; linear discriminant analysis; long-span latent topical information; natural language regularities; perplexity reduction; speech recognition; speech recognizer; statistical language modeling; topical mixture model; word error rate reduction; Adaptation model; Broadcasting; Computer science; Data mining; Error analysis; History; Large scale integration; Natural languages; Speech recognition; Vectors;
Conference_Titel :
Chinese Spoken Language Processing, 2004 International Symposium on
Print_ISBN :
0-7803-8678-7
DOI :
10.1109/CHINSL.2004.1409649