To expand CSR (continuous speech recognition) software to the mobile environmental use, we have developed embedded version of "Julius". Julius is open source CSR software, and has been used by many researchers and developers in Japan as a standard decoder on PCs. Julius works as a real time decoder on a PC. However further computational reduction is necessary to use Julius on a microprocessor. Further cost reduction is needed. For reducing cost of calculating pdfs (probability density function), Julius adopts a GMS (Gaussian Mixture Selection) method. In this paper, we modify the GMS method to realize a continuous speech recognizer on microprocessors. This approach does not change the structure of acoustic models in consistency with that used by conventional Julius, and enables developers to use acoustic models developed by popular modeling tools. On simulation, the proposed method has archived 20% reduction of computational costs compared to conventional GMS, 40% reduction compared to no GMS. Finally, the embedded version of Julius was tested on a developmental hardware platform named "T-engine". The proposed method showed 2.23 of RTF (Real Time Factor) resulting 79% of that of no GMS without any degradation of recognition performance.

No abstract

14 ms / frame -0.6 sec 1000We have developed speech recognition middleware on a RISC microprocessor which has robust processing functions against environmental noise and speaker ditkences. The speech recognition middleware enables developers and users to use a speech recognition process fbr m n y possible speech applications, such as car navigation system and handheld PCs. In this paper, we report implemntation issues ofspeech recognition process in middleware of microprocessors and propose robust noise handling hnctions using ANC(Adaptive Noise Cancellation) and noise adapt\ve models. We also propose a new speaker adaptation algorithm, in which the relationships among HMMs(Hidden Markov Models) transkr vectors are provided as a set of pre-trained interpolation coefficients. Experimental evaluations on 1000-word vocabulary speech recognition showed promising results h r both robust processing functions ofthe proposed noise handling methods and the proposed speaker adaptation method.

No abstract

This paper presents an OFDM transceiver for wireless LAN systems and its baseband transceiver architecture. TYPICAL PARAMETERS IN PACKET-ORIENTED OFDM SYSTEMS. We study the optimum parameters about DFT size, guard K DFT size interval, symbol duration, and number of subcarriers in an 80-Fb Bandwidth (Hz) MHz bandwidth by extending the IEEE802.11a standard. The T Guard interval (s) proposed transceiver has a maximum 300-Mbps transmit rate Tg DFT window length (s) and achieves 600 Mbps by use of a 4x2 MIMO system. We Tf OFDM frame length (s) have designed the SISO-OFDM transceiver in a 0.25-,um CMOS N8 Number of data subcarriers technology. The transceiver consumes about 800 mW at 2.5-V Nb Coded bits per subcarrier power supply and 80-MHz clock frequency. For verification, the R Coding ratearchitecture has been implemented to a FPGA prototype. We describe the parameters of our proposal and the TGn Sync optional proposal by comparison.