SPEECH RECOGNITION MODEL GENERATION DEVICE, METHOD FOR GENERATING SPEECH RECOGNITION MODEL, SPEECH RECOGNITION DEVICE, SPEECH RECOGNITION METHOD, SPEECH RECOGNITION SYSTEM AND RECORDING MEDIUM

摘要

PROBLEM TO BE SOLVED: To provide a speech recognition model generation device, which can generate speech recognition model reflecting the speech input characteristic of a speech recognition device, and to provide the generated speech recognition model to the speech recognition device. SOLUTION: The speech recognition device 1 is provided with a measurement section 11. The speech recognition model generating device 2 is provided with a reflecting data generation section 23 and a speech recognition model generating section 24. The filter coefficient measured by the measurement section 11 is transmitted to the speech recognition model generating device 2. The speech recognition model generating device 2 generates reflection data, reflecting the speech input characteristic of the speech recogniting device 1 from the received filter coefficient and sample data stored in a sample data storage section 21. Based on the generated reflection data, the speech recognition model generation section generates the speech recognition model.