Optional sentence pattern pronunciation inputting method and device

Abstract

Present invention relates to a speech input device. Said speech input device contains a sentence pattern selector unit for providing multi - sentence pattern, one output interface for outputting and switching said multi - sentence pattern for a user selecting, a sound identification unit for identifying said user inputted sound to obtain a identifying result, a content data base for storing data, and a data base search unit for searching corresponded data according to said identifying result. Present invention also provides a speech input method, which contains, providing multi - sentence pattern, displaying and said multi - sentence pattern, selecting one sentence pattern, starting up a pattern for corresponding said selected sentence pattern, inputting a sound, identifying said sound referring said pattern, generate a identifying result, inputting said identifying result to a data base searching unit and from said data base searching unit to a content data base to search said identifying resultant corresponded content.

Description

technical field [0001] The invention relates to a voice input method and a device, in particular to a voice input method and a device with a selectable sentence pattern. Background technique [0002] With the rapid development of speech recognition technology, the combination of speech recognition system and home appliances, communications, multimedia, information and other products is becoming more and more common. However, one of the problems often faced when developing speech recognition systems is that users do not know what to say when facing the microphone, especially if these products allow users a certain degree of freedom in speech input. Readers are often overwhelmed and unable to experience the benefits of using voice input. [0003] Currently, there are three types of voice input methods for devices with voice recognition function: [0004] 1. Provide a single sentence pattern input: the user can only input voice according to the single sentence pattern limited...

AI Technical Summary

Problems solved by technology

In addition, if natural language (Nature Language) is used as the input form, although the user is not restricted by sentence patterns at all, the error rate of speech recognition will also increase due to the large increase in the range of speech recognition.

[0006] 3. Provide a dialogue or similar dialogue mechanism: Under the guidance of the system interface prompts, the user completes the voice input action step by step between the system and the user. The disadvantage is that the whole process is easy to flow. It is lengthy, especially if the speech recognition often makes mistakes during the process, it will make the user lose patience

[0007] Since the above-mentioned three input methods have their inevitable defects, users cannot feel the benefits of using this natural and humanized interface when using an existing device with a voice recognition function, but will feel The use of this type of voice control device is not as good as manual button or keyboard input, so the application and promotion of voice control devices are also limited to a certain extent

Images