CN201845546U

CN201845546U - Device capable of controlling mobile phone through speech

Info

Publication number: CN201845546U
Application number: CN2010206135048U
Authority: CN
Inventors: 王新; 王林
Original assignee: XI'AN LONGFEI SOFTWARE CO Ltd
Current assignee: Shanghai Longcheer Technology Co Ltd
Priority date: 2010-11-18
Filing date: 2010-11-18
Publication date: 2011-05-25
Anticipated expiration: 2020-11-18

Abstract

The utility model discloses a device capable of controlling a mobile phone through speech, which comprises a speech recognition module, a serial peripheral interface (SPI) serial bus, a central processing module, a universal serial bus (USB) HOST bus, an interface socket, a speech recognition module power supply circuit, a central processing module power supply circuit, an interface power supply circuit and a power supply module. The speech recognition module is connected with the central processing module through an SPI serial bus interface. The central processing module is connected with the interface socket through the USB bus. The power supply module provides a power supply system for three modules and can charge the mobile phone and other terminal devices. The speech recognition module collects, recognizes and matches speech signals of a person and sends a processing result to the central processing module. According to a reported result, the central processing module operates a processing protocol and sends the processing protocol to the mobile phone through a USB interface. The utility model is convenient and is applicable to office work, vehicles and other situations in which manual mobile phone operation is not convenient.

Description

A kind of can voice the equipment of control mobile phone

Technical field

The utility model relates to electronic device field, relate in particular to a kind of can voice the equipment of control mobile phone, realize voice control by USB HOST interface to mobile phone.

Background technology

Speech recognition is a cross discipline.Recent two decades comes, and speech recognition technology is obtained marked improvement, begins to move towards market from the laboratory.Unspecified person speech recognition technology ASR (AutomaticSpeech Recognition), be that the language data of tens G is passed through the analysis of language specialist speech model, scientist sets up mathematical model, and the minutia of primitive voice is extracted in the process repetition training, and extract difference between each primitive, can obtain the minutia of the basic former sound on statistical probability optimization meaning, at last algorithm and speech model are converted into hardware chip, to be applicable in the various Embedded Application.

People estimate that in following 10 years, speech recognition technology will enter every field such as industry, household electrical appliances, communication, automotive electronics, medical treatment, home services, consumption electronic product.Speech recognition technology is chosen as one of ten major issues of computing machine development in 1997 in the application in some fields by US News circle.What add speech identifying function in present portable terminal is sub-fraction also, and most of terminal is owing to reasons such as cost, technology do not adopt this new technology.And the on-keyboard trend of portable terminal is more and more obvious, and interactive means is changing to more intelligent direction gradually also.

Summary of the invention

At the deficiency of prior art, the purpose of this utility model provide a kind of can voice the equipment of control mobile phone.

The technical scheme that the utility model designing institute adopts is:

A kind of can voice the equipment of control mobile phone, comprise: sound identification module, SPI universal serial bus, central processing module, USB HOST bus, interface socket, sound identification module feed circuit, central processing module feed circuit, interface feed circuit, supply module, described sound identification module is connected by serial line interface with central processing module, and central processing module operation such as reads and writes by serial line interface to sound identification module; Central processing module is connected by usb bus with interface socket; Supply module is sound identification module, central processing module, interface socket power supply by sound identification module feed circuit, central processing module feed circuit, interface feed circuit respectively.

Described sound identification module is by Mike, and voice recognition chip is formed, and unspecific people's sound carried out A/D, D/A conversion, and order is encoded, and is sent to central processing module.Sound identification module is gathered peripheral voice command by Mike, and sends to the voice recognition processing chip by mimic channel.

Described employing has the chip of USB host function as central processing unit, and its effect is the voice command that the processed voice identification module sends over, and after parsing, is undertaken finishing the control to mobile phone alternately by USB interface and mobile phone.It is to set up a cover interaction protocol that this equipment carries out mutual basis with mobile phone, and can download in the mobile phone by USB interface.When mobile phone is connected to this equipment, this equipment can be automatically to mobile phone-downloaded, install and drive, to realize mutual with mobile phone.

In the equipment of described voice control mobile phone, the coupling of voice command is to pre-set some voice commands, and order that collects and the order that sets in advance are mated, if matching result is arranged, the result is sent to central processing module by interaction protocol.

In the equipment of described voice control mobile phone, interaction protocol is meant that sound identification module and central processing module carry out data interaction, realizes exchanges data by the SPI interface between the two.

In the equipment of described voice control mobile phone, central processor unit is by processor, compositions such as clock, and the order of control speech recognition sends terminal devices such as mobile phone to.

In the equipment of described voice control mobile phone, processor is a chip that has SPI interface, USB HOST interface at least, resolves the voice command that sound identification module sends over, and sends to terminal device such as mobile phone by USB interface.

In the equipment of described voice control mobile phone, USB interface is the approach of this equipment and mobile communication.Both carry out the transmission and the parsing of order by default interaction protocol.

Description of drawings

Fig. 1 is the utility model principle mount structure synoptic diagram;

Fig. 2 is the structural representation of the utility model speech recognition;

Fig. 3 is the utility model sound identification module and central processing module communication scheme;

Fig. 4 is the utility model central processing module and socket connection diagram.

The number in the figure explanation:

1-sound identification module 2-SPI universal serial bus 3-central processing module

4-USB bus 5-interface socket 6-sound identification module feed circuit

Core 12-mimic channel is handled in the 7-central processing module feed circuit 8-interface feed circuit 9-supply module 10-Mike 11-of system identification

Specific implementation method

How further specify the utility model below in conjunction with accompanying drawing realizes:

As shown in Figure 1, be connected by parallel interface or serial line interface 2 between sound identification module 1 and the central processing module 3, central processing module 3 carries out read-write operation by SPI universal serial bus 2 interfaces to sound identification module 1; Be connected by usb bus 4 between central processing module 3 and the interface socket 5; Supply module 9 is sound identification module 1, central processing module 2, interface socket 5 power supplies by sound identification module feed circuit 6, central processing module feed circuit 7, interface feed circuit 8 respectively.

As shown in Figure 2, sound identification module 1 is by Mike 10, voice recognition chip is formed, sound to unspecific people carries out A/D, D/A conversion, and order encoded, be sent to central processing module 3, gather peripheral voice command by Mike 10, and send to voice recognition processing chip 11 by mimic channel 12.Finish speech recognition and mainly pass through following steps:

(1) initialization system state

(2) the typing voice command writes recognized list.It is that maximum can be discerned clauses and subclauses that chip has individual parameter, as, certain chip is supported 50 clauses and subclauses, each clauses and subclauses is Chinese phonetic alphabet of standard mandarin, between per two words with a space.Notice that the clauses and subclauses of typing herein are keyword not all.The numbering of the entry of typing can be identical, and the numbering of for example " making " with " dialing " two keywords is the same.When the people sent the voice command of " making " XXX or " dialing " XXX, the numbering that can return these two keywords after the speech recognition success got final product.

(3) begin identification, can begin speech recognition by control chip by certain register that changes chip.

(4) interrupt response, Mike can be sent to chip with the voice signal that collects.The voice content that chip can send the user is a phonetic feature by spectral conversion, then and the key term of typing in advance mate, a result of Optimum Matching is as recognition result.No matter need what content identified is,, the entry content of needs identification will be sent to recognition engine with character style as long as revise related register.

Realization about speech recognition mainly contains dual mode: interrupt mode and polling mode.Whether interrupt mode is meant as long as Mike collects sound, no matter be voice command, all can produce interruption; Poll then is to set one to trigger password, for example " begins identification ", and implementation is as follows: the state that is in a circulation identification during equipment normality, behind the voice command that recognizes " beginning identification ", at this moment, equipment can provide prompt tone, and the prompting user said voice command in several seconds.

As shown in Figure 3, sound identification module 1 and central processing module 3 communicate by spi bus.Central processing module 3 is by some register of spi bus read-write voice recognition chip, and promptly the may command sound identification module is discerned the user's voice order, perhaps reads recognition result.

As shown in Figure 4, central processing module 3 is connected with socket 5 by usb bus.After mobile phone connected by the USB line, this moment, mobile phone can be identified as a SLAVE equipment of this equipment.This equipment is communication protocol with the basis that mobile phone is finished communication interaction, and this communication protocol has comprised the parsing of order, the judgement of state etc.For example, central processing module will send a command to mobile phone by usb bus after reading the order of calling from sound identification module, and mobile phone starts the program of calling and opens hand-free function simultaneously according to communication protocol.

Claims

One kind can voice the equipment of control mobile phone, comprise: sound identification module, SPI universal serial bus, central processing module, USB HOST bus, interface socket, sound identification module feed circuit, central processing module feed circuit, interface feed circuit, supply module, it is characterized in that: described sound identification module is connected by the SPI serial bus interface with central processing module, and central processing module is connected by usb bus with interface socket.
2. according to claim 1 a kind of can voice the equipment of control mobile phone, it is characterized in that: described supply module is sound identification module, central processing module, interface socket power supply by sound identification module feed circuit, central processing module feed circuit, interface feed circuit respectively.