Controller general type speech play control system for identifying
Technical field
The utility model relates to a kind of control system, relates in particular to a kind of SCM Based control system, is used to control the broadcast and the identification of voice.
Background technology
Along with electronic technology and computer technology rapid development, household electrical appliance and the hommization day by day of other electronic product, easy to use, simple to operate.People with their information interchange in, voice are the most direct, convenient and natural mode.They have had " ear " seemingly, and you are as long as open one's mouth, and just understand your intention and by the instruction behaviour.
Traditional embedded speech system mainly contains: 1 application-specific integrated circuit that is made of multi-channel filter and linear matched; 2 voice systems of forming by DSP; 3 application-specific integrated circuits add AD conversion, DA conversion and storer etc.But they have shortcoming separately, low as: phonetic recognization rate, cost an arm and a leg, system complex etc.And the flow process of whole broadcast identification control program is very complicated, causes programming and upgrading difficulty, and a large amount of uses of embedded speech system are restricted.
The utility model content
The purpose of this utility model provides a kind of controller general type speech play control system for identifying, and simplied system structure to reduce cost, to shorten the construction cycle, strengthens expandability.
A kind of controller general type speech play control system for identifying of the present utility model comprises singlechip controller, speech play module, sound identification module and speech control module.Speech play module, sound identification module and speech control module are connected with singlechip controller respectively.
Sound identification module is sent to singlechip controller to the voice signal after discerning, by singlechip controller the voice signal that receives being passed to speech control module handles, voice signal after the processing is being sent to the speech play module through singlechip controller, and feeds back to sound identification module.
The beneficial effect that the utility model is realized:
A kind of controller general type speech play control system for identifying described in the utility model comprises singlechip controller, speech play module, sound identification module and speech control module.Speech play module, sound identification module and speech control module are connected with singlechip controller respectively.The described controller general type of utility model speech play control system for identifying, have simple in structure, cost is low, the characteristics of easy realization, can shorten the construction cycle, and expandability is strong.
Description of drawings
Fig. 1 is the synoptic diagram of the utility model speech play control system for identifying one embodiment.
Embodiment
Below describe the technical solution of the utility model in detail.Only for the explanation concrete grammar, its scale of this method is not subjected to the restriction of embodiment to the utility model embodiment.
Embodiment
As shown in Figure 1, controller general type speech play control system for identifying described in the utility model comprises singlechip controller 1, speech play module 2, sound identification module 3 and speech control module 4.Speech play module 2, sound identification module 3 and speech control module 4 are connected with singlechip controller 1 respectively.
Voice signal after 3 pairs of identifications of sound identification module is sent to singlechip controller 1, by singlechip controller 1 voice signal that receives being passed to speech control module 4 handles, voice signal after the processing is being sent to speech play module 2 through singlechip controller 1, and feeds back to sound identification module 3.
In the present embodiment, the SPCE061A type controller that singlechip controller 1 adopts Sunplus Technology Co., Ltd to produce is controlled speech play module 2, sound identification module 3 and speech control module 4.
In the present embodiment, the SPCE061A that sound identification module 3 adopts Sunplus Technology Co., Ltd to produce, finish to voice identification.The speech recognition algorithm principle is: in the training stage, the user gives an account of the speech in the vocabulary successively, extracts the proper vector of each frame sampling data, and proper vector is deposited in the template base as template.At cognitive phase, extract the proper vector wait to know voice, then with its successively with template base in each template carry out similarity relatively, the soprano as a result of exports with similarity.
In the present embodiment, the SPCE061A that speech control module 4 adopts the Sunplus Technology Co., Ltd of 32 I/O ports to produce is to operate accordingly after the speech recognition success.After the single-chip microcomputer speech recognition success, can send signal to the external world by the I/O port and successfully instruct, finish the function of whole system to trigger external action or voice broadcast.Its each pin all can pass through the software simple operations again, is defined as separately and inputs or outputs mouth, is used to recognize behind the special sound to send functions such as signal, outer down trigger mouth, input signal mouth to the outside.
In the present embodiment, speech play module 2 is finished speech play.Speech processes roughly can be divided into encoding process, storage, decoding processing and A/D conversion.Therefore with database storage, decoding and D/A etc. are made corresponding playing module, and it is optimized compression, only need to understand the function that this module will realize and the content of corresponding advanced interface api function and parameter thereof, call this api function then and can realize this function, the very big like this software that made things convenient for generates the respective streams range code automatically.