An ARM-based system-on-a-programmable-chip architecture for spoken language translation

Shun Chieh Lin, Jia Ching Wang

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

Previous research shows that there are two architectures for spoken language translation (SLT) system implementation. One is client-server based systems that should be built on the server computer but unreliability of the remote connection. The other is to build portable stand-alone devices but it lacks real-time performance. In this brief, a system-on-a-programmable-chip (SoPC) solution is proposed by realizing the entire SLT system within a single chip. This SoPC is characterized by small size, low cost, real-time operation, and high portability. This entire design was implemented on ALTERA EPXA10 device. Performance for English-to-Mandarin translation process can be completed within 1 s at a 46.22-MHz clock frequency with 3000 translation patterns. The total logic usage of the EPXA10 device is 50% (about 19318 logic cells).

Original languageEnglish
Pages (from-to)765-769
Number of pages5
JournalIEEE Transactions on Circuits and Systems II: Express Briefs
Volume54
Issue number9
DOIs
StatePublished - Sep 2007

Keywords

  • ARM
  • language translation
  • speech processing speech recognition
  • system-on-a-programmable-chip (SoPC)

Fingerprint

Dive into the research topics of 'An ARM-based system-on-a-programmable-chip architecture for spoken language translation'. Together they form a unique fingerprint.

Cite this