CN2776505Y - Sound control device of toy - Google Patents

Abstract

The utility model provides a sound control device for a toy, which includes a sound pickup module and a sound playback module respectively connected to a main control circuit board. The main control circuit board includes a control module, a voice recognition module, a voice synthesis module, a model and a rule library module, A/ The D conversion module, the D/A conversion module, the pre-amplification module, the power amplification module and the control module are respectively connected with the computer and the execution module through the input interface module and the output interface module. The utility model allows the toy to distinguish and respond to the user's speech, and is highly interesting. The updated model and rule library can be downloaded and updated through a computer, and can be transplanted to other toys, and different sounds can also be adjusted.

Description

A toy voice control device

Technical field

The utility model relates to the technical field of voice control toys, in particular to a voice control device for toys.

Background technique

With the continuous improvement of living standards, the variety of toys is becoming more and more abundant, and people's requirements for toys are also getting higher and higher. Traditional toys mainly attract children through appearance, color, sound, intellectual exercise, etc. Most of them need to be "held" by hand to play, such as building blocks, dolls, toy cars, etc., which need to be operated by hands. These toys cannot "listen" to understand and "answer" the user's speech. When children use them, they can only rely on their own unilateral operation and imagination. It is difficult to obtain the same toys as their favorite toys (especially anthropomorphic toys, Such as dolls, etc.) to "communicate" the feeling. In recent years, many toys have added voice control or sounding functions, so that they can respond to people's voices, such as walking in the direction of the sound, or saying something or making certain sounds after certain operations. Sound, etc., but the shortcomings of this type of toy are: (1) the function is simple, the existing sound or voice-activated toys are still unable to distinguish the content of the speech, there is still a long way from the feeling of "communication", and the interest is low; ( 2) Toy usage rules and program content cannot be updated. Existing toys cannot change their usage rules after the production is completed. For example, some toys that can make sounds can only say the same few words repeatedly after the production is completed. It is easy to feel boring when it is long, it is difficult to satisfy children's curiosity, and the playability is low.

Contents of Invention

The purpose of the utility model is to overcome the shortcomings and deficiencies of the above-mentioned prior art, and provide a toy voice control device that is interesting, highly playable, and capable of voice interaction and control and program update.

The purpose of the utility model is achieved through the following technical solutions: the toy sound control device includes a main control circuit board, a sound pickup module, and a sound playback module, and the sound pickup module and the sound playback module are respectively connected to the main control circuit board, and it is characterized in that : the main control circuit board includes a control module, a speech recognition module, a speech synthesis module, a model and a rule base module, an A/D conversion module, a D/A conversion module, a preamplifier module, and a power amplifier module, and the control module Connect with speech recognition module, speech synthesis module, model and rule storehouse module respectively, described speech recognition module, speech synthesis module respectively pass A/D conversion module and preamplification module, D/A conversion module and power amplifier module and pick-up The sound module and the sound playback module are connected, and the speech recognition module and the speech synthesis module are connected with the model and the rule base module at the same time, and the control module is also connected with the computer and the execution module through the input interface module and the output interface module respectively.

In order to better realize the utility model, the control module, speech recognition module and speech synthesis module are integrated by a microprocessor.

The sound pickup module may be a microphone.

The sound playing module may be a loudspeaker.

The execution module may be a light, a display screen or a transmission device of the toy.

The control module, speech recognition module, and speech synthesis module can also be independent control chips, speech recognition chips, and speech synthesis chips. The model and rule base module can be memory, and the input interface module and output interface module can be It is a conventional IO interface, the A/D conversion module and the D/A conversion module may be A/D converters and D/A converters, the preamplifier module may be a preamplifier circuit, and the The power amplification module may be a power amplifier circuit.

The software part of the model and rule library module, that is, the model and rule library includes a speech recognition model library, a speech synthesis model library, an output control signal library, and a rule library. The speech recognition model library includes all vocabulary recognition models that enable the device to generate responses, the speech synthesis model library includes all synthetic models of speech that the device can produce, and the output control signal library includes all that enable the device The control signal that can be generated, the rule base contains all mapping rules between speech recognition models, speech synthesis models, and output control signals.

The method for realizing voice interaction and control of the toy voice control device of the utility model includes:

(1) convert the sound wave into an analog electrical sound signal by the pickup module, and transmit it to the A/D conversion module after being amplified by the preamplifier module;

(2) described A/D conversion module carries out sampling and quantization to the sound electric signal of simulation, is converted into digital signal and sends into described speech recognition module;

(3) The speech recognition module detects and recognizes the input signal in real time, and if the input signal contains vocabulary in the speech recognition model library, the recognition result is sent to the control module, otherwise the input signal is continued detection;

(4) After the control module receives the recognition result provided by the speech recognition module, according to the predetermined reaction rules in the rule base, output the required control to the output interface module, which is pre-defined in the output control signal base signal, and notify the voice recognition module to synthesize the required voice;

(5) according to the requirements of the control module, the speech synthesis module utilizes the speech recognition model in the speech recognition model library to synthesize the required speech signal, and sends it to the D/A conversion module;

(6) Utilize the D/A conversion module to convert the digital voice signal output by the speech synthesis module into an analog electrical voice signal, and play the analog voice signal through the power amplification module and the sound playback module.

The method for realizing the program update of the toy voice control device of the utility model comprises:

(1) The device continuously detects whether the input interface module is connected to the computer. If it detects that the input interface module is connected to the computer, it stops all operations and enters the program update state;

(2) When the device is in the program update state, download and update the model and rule base through the computer;

(3) When the download of the model and rule base is completed, and the computer is disconnected from the input interface module, the device returns to the normal working state, and realizes voice interaction and control by using the new model and rule base.

Compared with the prior art, the utility model has the following advantages and beneficial effects:

(1) Strong interest, the device can make the toy distinguish and respond to the user's speech, create a more realistic anthropomorphic feeling, and make the toy more convenient to operate and more interesting.

(2) It can be updated repeatedly. The device can download and update the model and rule base through the computer, so that the same toy can have different speech contents and reaction modes, which increases the playability of the toy.

(3) Portability, the device can be easily transplanted to different types of voice-activated toys by setting appropriate vocabulary and rule bases.

(4) Adjustability, the device can easily synthesize sounds with different characteristics (such as male voice, female voice, child voice, etc.) , or achieve different speaking effects in the same toy.

Description of drawings

Fig. 1 is the circuit block diagram of the toy sound control device of the present invention.

2 and 3 are schematic circuit diagrams of the toy voice control device of the present invention.

Fig. 4 is an installation structure diagram of the toy voice control device of the present invention.

Fig. 5 is a working flow chart of the toy voice control device of the present invention.

Detailed ways

The utility model will be described in further detail below in conjunction with the embodiments and accompanying drawings, but the implementation of the utility model is not limited thereto.

As shown in Figure 1, the toy voice control device comprises a main control circuit board 1, a sound pickup module 2, and a sound playback module 3, and the sound pickup module 2, and the sound playback module 3 are respectively connected with the main control circuit board 1, and the main control circuit board 1 Including control module 4, speech recognition module 5, speech synthesis module 6, model and rule library module 7, A/D conversion module 8, D/A conversion module 9, input interface module 10, output interface module 11, preamplification module 12, power amplification module 13, control module 4 is connected with speech recognition module 5, speech synthesis module 6, model and rule storehouse module 7 respectively, speech recognition module 5, speech synthesis module 6 pass A/D conversion module 8 and front end respectively Amplifying module 12, D/A conversion module 9 and power amplifying module 13 are connected with pickup module 2, playback module 3, and speech recognition module 5, speech synthesis module 6 are connected with model and rule storehouse module 7 simultaneously, control module 4 It is also connected to the computer 14 and the execution module 15 through the input interface module 10 and the output interface module 11 respectively. Wherein the sound pickup module 2 of the toy sound control device is a microphone, the sound playback module 3 is a loudspeaker, and the execution module 15 can be a light, a display screen or a transmission device of the toy.

As shown in Figures 2 and 3, the control module 4, speech recognition module 5, and speech synthesis module 6 of this toy voice control device are microprocessors that combine the three into one, which is a DSP chip TMS320F2812, and the model and rule base module 7 are composed of TMS320F2812 The built-in FLASH memory is formed on-chip. The A/D conversion module 8 is composed of the ADC in the TMS320F2812 chip and the low-pass filter circuit outside the chip. The D/A conversion module 9 is composed of the PWM circuit in the TMS320F2812 chip and the on-chip The external low-pass filter circuit is jointly formed. The input interface module 10 and the output interface module 11 are realized by the IO port provided in the TMS320F2812 chip. The preamplifier module 12 is composed of a preamplifier circuit, and the power amplifier circuit 13 is composed of a power amplifier circuit. The sound pickup module 2 adopts a microphone, and the sound playback module adopts a loudspeaker.

As shown in Figure 4, when the toy voice control device is installed on a toy dog, the main control circuit board 1 comprising DSP chip TMS320F2812, preamplifier circuit, power amplifier circuit and low-pass filter circuit is placed in the toy dog shell, The microphone and the loudspeaker are respectively placed on the head and torso of the toy dog, and are connected with the preamplifier circuit and the power amplifier circuit on the main control circuit board 1 through the signal line; One IO port of TMS320F2812 on the circuit board 1 is connected; the inside of the shell contains the transmission device of the toy dog's limbs, head and tail, and the other IO port of the TMS320F2812 on the main control circuit board 1 provides control signals for the transmission device.

This toy voice control device is controlled by software to achieve the purpose of voice interaction and control and program update. Figure 5 shows the workflow of this toy voice control device, as shown in Figure 5:

When the power is turned on, the initialization work is performed first, and then the main program cycle is entered.

In the main program cycle, it first checks whether the input interface module is connected to the computer. When it is detected that the input interface module is connected to the computer, it enters the program update state. At this time, all normal operations are stopped, and the user controls the model and the computer through the computer. The rule base is updated. After the update is completed, the user disconnects the computer from the input interface module, and the system returns to a normal working state. The new model and rule base are used for voice interaction and control with the user. If it is not detected that the input interface module is connected to the computer, it waits for the input buffer to be full before entering the speech recognition module of the soft implementation.

In the speech recognition module, the following steps are used to detect and recognize the input signal:

(1) Use the endpoint detection technology in speech recognition to detect whether the input signal contains speech. If it is detected that the input signal contains speech, then find out and record the starting point and end point of the speech; if it is not speech, then return to the beginning of the main program loop and wait for new data input.

(2) Using speech recognition and recognition rejection technology, the speech segment recorded in step (1) is compared with the speech model in the speech recognition model library, if the speech is matched with one of the models, then the speech is The system can recognize vocabulary. At this time, write down the serial number of the model, and send the recognition result to the control module; if the speech does not match all the models, the speech is not a system-recognizable vocabulary, and return to the main program loop. Start to continue execution, waiting for new data input. In the speech recognition model library, the speech recognition model uses Speech recognition model serial number speech recognition model stored in the form.

After receiving the serial number of the speech recognition model provided by the speech recognition module, the soft-implemented control module adopts the following steps to determine the output control signal and the speech to be synthesized:

(1) Find the corresponding response rules in the rule base according to the recognition results of the speech recognition module. The rule base is a table that records the mapping between the serial number of the speech recognition model, the serial number of the speech synthesis model, and the serial number of the output control signal, and its format is: Speech recognition model serial number Speech synthesis model serial number output control letter serial number , for example, the serial number of the speech recognition model representing the name of the toy is 1, the serial number of the speech synthesis model responding to the voice "Wow" is 3, and the serial number of the output control signal of the response action "Wag the tail" is 7, then the form of the rule table is 1 3 7 , when the name of the toy is called, the speech recognition module can obtain the speech recognition model serial number as 1, and find the corresponding speech synthesis model serial number and output control signal serial number as 3 and 7 from the rule table.

(2) Find the corresponding output control signal in the output control signal library according to the obtained serial number of the output control signal, and output the signal to the output interface module. In the output control signal library, the output control signals are stored in the following format: Output control signal serial number output control signal .

(3) Send the synthesized voice sequence number obtained to the voice synthesis module realized by the software.

In the speech synthesis module, the desired speech is synthesized using the following steps:

(1) According to the serial number of the speech synthesis model provided by the control module, search the corresponding speech synthesis model in the speech synthesis model library. In the speech synthesis model library, the speech synthesis model uses Speech Synthesis Model Sequence Number speech synthesis model stored in the form.

(2) Use speech synthesis technology to synthesize the desired speech and output it to the D/A converter.

After completing these tasks, return to the main program loop and start to execute again.

As mentioned above, the utility model can be better realized.

Claims (5)

1, a kind of toy sound-control device comprises main control board, pickup module, playback module, described pickup module, playback module is connected with main control board respectively, it is characterized in that: described main control board comprises control module, sound identification module, the phonetic synthesis module, model and rule base module, the A/D modular converter, the D/A modular converter, pre-amplifying module, power amplifier module, described control module respectively with sound identification module, the phonetic synthesis module, model is connected with the rule base module, described sound identification module, the phonetic synthesis module is respectively by A/D modular converter and pre-amplifying module, D/A modular converter and power amplifier module and pickup module, playback module connects, and sound identification module, the phonetic synthesis module is connected with the rule base module with model simultaneously, and described control module is also passed through input interface module respectively, output interface module and computer, Executive Module connects.

2, by the described a kind of toy sound-control device of claim 1, it is characterized in that: described control module, sound identification module, phonetic synthesis module are integrated by a microprocessor.

3, by the described a kind of toy sound-control device of claim 1, it is characterized in that: described pickup module comprises microphone.

4, by the described a kind of toy sound-control device of claim 1, it is characterized in that: described playback module comprises loudspeaker.

5, by the described a kind of toy sound-control device of claim 1, it is characterized in that: described Executive Module comprises light, display screen or the transmission device of toy.

Images