With the popularity of smart home appliances, more and more products no longer need to adjust and control manually. Voice awakening and handling are more convenient for people's lives. I would like to ask, not only the wake -up service of the smart butler, but also the time to remind you, is it more comfortable than getting up in a busy getting up?

At the same time as a comfortable life, have you considered a voice control system composed of voice collection modules, pre -level processing modules, voice training modules, voice recognition modules, voice prompt modules, and output control modules. What functions are there? Let's answer them one by one. The voice acquisition module mainly completes the signal conditioning and signal acquisition functions. It converts the original voice signal converting idiom vein sequence. Therefore, the module mainly includes signal processing processes such as sound/electrical conversion, signal conditioning and sampling. The main function of the pre -level processing module is to filter the interference signal, extract the voice characteristics vector, and turn the extracted voice characteristics vector quantity to the standard voice feature vector.

  Therefore, this module mainly includes voice signal processing processes such as voice processing, feature extraction, vector quantification. The main function of the voice training module is to count the probability of the pronunciation standard vector collected and extracted multiple times, extract the speaker's best voice characteristics standard vector to prevent the extraction feature parameters caused by the speaker's mood and environmental factor and other factor parameters. The voice recognition effect, so this module mainly includes the processing process of probability statistics and parameter assessment, and is implemented with the Hidden Marcov model (HMM model). The main function of the voice recognition module is to compare the re -collected standard voice feature vector with the voice model in the voice template library to determine the current voice command function. Therefore, the module mainly includes two processes: vector comparison and parameter assessment. The main function of the voice prompt module is to prompt the user to perform related operations or instructions based on the results of voice recognition. Therefore, the module mainly includes the calling process of prompting voice resource files, D/A conversion, signal amplification and other voice processing processes. The main function of the output control module is to output the corresponding control signal according to the results of the voice recognition, and realize the voice control function of the office appliances such as electric lights, television, and fans. Therefore, this module mainly includes signal drivers, output controllers and control objects. The main function of the voice template library is to store the best standard voice feature vector after training. The artificial intelligence of Xiaowei housekeeper under the world. Xiao Wei, an artificial intelligence product under Rong Sheng, is a smart chip with voice recognition and control. A "voice key" can complete all the platform selection, chasing drama, watching or playback, listening, translation, tuning, playing games, returning homepage and other operations on the TV's large screen. Not only can help children learn, but they can also help the elderly to relieve their boredom. Do not use a row of dense remote control buttons, just hold the "voice key", say the channel you want to see, and then enjoy the fun in life.