tags: raspberry pie Single-chip microcomputer stm32 51 single-chip microcomputer
Model: YS-LDV7
Name: Integrated Voice Recognition Module
Specifications: 43 * 29.7mm
Power supply voltage: 5V (internal operating voltage 3.3V)
Stander current: 30mA
Identify time current: 45mA
IO port output: high level is 3.3V
Communication method: serial communication (5V TTL level, can not directly pick RS232, RS485)
Microcontroller parameters: Model -> STC11L08XE, Flash-> 8K, SRAM-> 1280, EEPROM-> 32K
Identification entitlement: 50 sentences
Work temperature: -20 to 60 ° C
The actual principle of this module is a voice recognition module formed by 1 STC11 single-chip +1 LD3320 combination. The voice recognition section has been written, which only needs to program the STC microcontroller to add its own recognition statement and control program. The speech recognition part does not require ingredients and processes. This user only needs to make a single-chip programming application on the provided source program. This module has 1 5V TTL serial port and 16 microcontroller IO ports, and can communicate interactive information with external single-chip microcomputers. equipment.

The serial port of this module is shown in the following figure, from the bottom to 5, respectively, GND, RXD, TXD, 5V, 3.3V, where RXD and TXD are serial port reception and transmit ports of the actual internal microcontroller, so connect to external serial devices. Pay attention to the cross-pick up of RXD connecting TXD. The 3.3V port is the output port, which can be used for the supply utilization of external devices, generally not directly ignored directly. The picture below shows the connection method of the USB turn TTL and the voice module provided by the store. When connecting to other microcontrollers or other devices, pay attention to the serial level level required to be 5V, which is consistent with the speech module level, such as the level is inconsistent, can be added to the serial level The conversion module does not result in an abnormal communication exception or even damage the device port.

Open the program engineering, find the uint8 ld_asraddfixed () function in the ldchip.c file, and you can find the following figure below in this function:

-Srecog [] [] array is a keyword array, adding content as pinyin input mode, such as wanting to add "Lights" commands, write "kai deng", pinyin between each Chinese character separated by space. -Pcode [] array is the identification code array, the added identification code is a predetermined macro definition constant value, and must correspond to a keyword, as shown in the figure, the "da ma ce shi" command corresponds to the identification code For Code_DMCs.
Open the program, find the following figure below in the ldchip.h file:

- This is an addition and modification of the identification code, and the user can arbitrarily define the identification code and macro name according to its own needs and preferences, but must be paired with the identification code used earlier, otherwise the error will not be defined. The parameter range of the codeword is 01-FF, there is no specific meaning, as long as there is no repetition.
Open the program, find the void user_handle function in the main.c file, you can see the following figure in this function:

- The user can add the operation after the identification success is added to the corresponding identification code according to its own usage. Then, the corresponding action will be performed after the device to a sentence. In fact, this program is judged to identify that identification code and perform the corresponding action, which is a single-chip program application processing section.
This module program download is actually downloading the STC microcontroller, first we need to install USB to TTL driver (if installed without installing), then then USB turn TTL. 1. Open "STC-ISP Download Software", indicate all options according to the arrows of the figure below.

(Note: The single-chip model is in the series of STC11F60XE; the serial number is not necessarily COM4, what is the word is selected according to the software identification, which one is word, open the program file, open the source program -Obj folder, the suffix is. HEX File) 2, finally we first unplug the GND of the voice module, then click the Download / Program button, then plug in the GND, this is the progress bar display at the bottom of the software, will be prompted to complete. 3. If the download does not respond host, check the wiring method again, and whether the serial number is selected correctly.
Find the following microphone volume definition in the ldchip.h file

According to the scope of the description, the maximum sensitivity recognition distance can reach 5 meters (the ideal distance in the quiet small room), and the smallest sensitivity is suitable to be shouting near the mouth, and the sound interference to the long distance is very small. Adjust appropriately according to your usage. In order to improve the anti-interference ability, it is recommended to use the password identification mode program (default), and pay attention to modify the password (default is "Xiaojie"), the modification of the password is recommended in 3-8 words and modify the relatively remote words, which can provide identification Rate and reduce misunderstandings.
If you want to call Xunfei voice recognition module flexibly, you need to understand the function of each function, as well as the calling steps and methods. After watching the Xunfei speech recogniti...
When the face recognition module is not used in the implementation project, the context of the conversation can be controlled through the startConversation function and the resetConversation function,...
table of Contents Speech Recognition Mel frequency cepstral coefficient (MFCC) matrix Speech Recognition Mel frequency cepstral coefficient (MFCC) matrix Firstly, the audio input is divided into sever...
1. The speech synthesis is the text to speech, speech to text recognition is. 2. When looking at the uni-app framework demo, found that there are built-in voice recognition module, naturally think of ...
Speech recognition technology is also called automatic speech recognition. The goal is to automatically convert human speech content into corresponding text or text into speech by a computer. Text to ...
table of Contents Speech Recognition The history of speech recognition The processing flow of speech recognition tasks Speech recognition applications Speech Recognition The technology that machines c...
At the 2016 WWDC, Apple introduced a very useful speech recognition API, the Speech framework. The Speech framework can help you quickly integrate the voice input function into the APP. 2016 WWDC Spee...
I bought 2 voice modules, a Longqiu SYN7318, an offline voice module of Xunfei, and today I recorded the transplant SYN7318. This is a program that has been ported to the punctual atom F1 development ...
Platform: RK3399 OS: Android 7.1 Kernel: v4.4.83 demand: Use the Spirent solution to do speech recognition without affecting the normal microphone recording function. analysis: Since the speech is als...
Wherein the non-speech recognition module 51 embedded microcontroller Click to download the code and supporting information Download video entry (1) according to FIG (P4 interface) in FIG. 1, download...