Speech Recognition/Synthesis Core Delivers High Accuracy

Consistent accuracy in the range of 97% to 100% (typically higher than 98%) in multiple languages, including English, French, German, Dutch, Hebrew and Chinese, is promised by the firm’s C-language speech recognition and synthesis (SRS) core. The core has applications in mobile phones, remote controls, automotive climate control, toys and other systems with a voice-controlled user interface.It’s offered in object code for DSP or RISC processors and PC platforms; as a VHDL or Verilog core, with or without interfaces to other on-chip logic; as a cell-based SOC including codec and amplifiers; or in a complete OEM module that includes speaker, microphone, IFR, RF and other functionalities. The core currently runs on DSP processors from Texas Instruments and National Semiconductor; it can be compiled to run on any DSP or RISC machine. Object codes are available for a per-chip license fee.