CN211557436U - External audio input device - Google Patents

Abstract

The utility model provides an external audio input equipment, including the microphone of the multiple different models that can pick up same sound signal respectively. The utility model discloses a set up the microphone of multiple different models in audio input equipment, can acquire the output signal of different microphones under same sound source to can train speech recognition algorithm based on the multiple different output signal of same content, improve algorithm adaptability, realize the adaptation to the microphone sensor of different brand models.

Description

External audio input device

Technical Field

The utility model relates to an audio acquisition and transmission technical field especially relate to an external audio input equipment.

Background

With the rapid development of voice communication technology, in an era where intelligent electronic devices such as intelligent vehicles, intelligent homes, robots and wearable devices are becoming hot, voice interaction becomes the first choice for human-computer interaction due to convenience. Generally, the experience of the user using the intelligent terminal can be greatly improved through voice recognition or speaker identity recognition, so that the intelligent terminal with the voice recognition or speaker recognition function is bound to become a trend in the future. However, the speech recognition algorithm or the speaker recognition algorithm often depends on the signals picked up by the microphone on the terminal, and because of the difference in technical and technological standards of manufacturers of microphone sensors, the signals picked up by the microphone have a great relationship with the brand and model of the microphone sensor used, so that the recognition effect of the speech recognition algorithm may be different when different audio devices are matched, thereby resulting in poor user experience of the speech recognition algorithm or the speaker recognition algorithm.

SUMMERY OF THE UTILITY MODEL

The utility model aims to solve the technical problem that a thereby set up the microphone of multiple different models simultaneously and pick up different audio signal's external audio input equipment to same kind of sound is provided.

The utility model discloses a solve above-mentioned technical problem through following technical scheme: an external audio input device includes a plurality of different types of microphones capable of picking up the same sound signal, respectively.

The utility model discloses a set up the microphone of multiple different models in audio input equipment, can acquire the output signal of different microphones under same sound source to can train speech recognition algorithm based on the multiple different output signal of same content, improve algorithm adaptability, realize the adaptation to the microphone sensor of different brand models.

Preferably, the device comprises a shell, wherein a main circuit board, a USB interface and a microphone circuit board which are in communication connection with the main circuit board, and an AV interface for collecting reference sound signals are arranged in the shell, and the microphone is welded on the microphone circuit board; the USB interface and the AV interface are arranged on the surface of the shell.

Preferably, the AV interface is in communication connection with a microphone circuit board or a main circuit board.

Preferably, still be provided with the vice microphone circuit board with main circuit board communication connection in the casing, the even array welding of edge circumferencial direction has a plurality of main microphones on the microphone circuit board, and the welding has the vice microphone of a plurality of different models on the vice microphone circuit board, the AV interface still with vice microphone circuit board communication connection.

Preferably, the microphone circuit board and the auxiliary microphone circuit board are parallel to each other and fixed inside the casing, the main circuit board is fixedly matched with the casing and limited between the microphone circuit board and the auxiliary microphone circuit board, the auxiliary microphone circuit board is arranged below the main circuit board, and the USB interface and the AV interface are fixed below the auxiliary microphone circuit board.

Preferably, the main circuit board is further connected with an interface circuit board through an insertion pin plate.

Preferably, the microphone circuit board, the auxiliary microphone circuit board and the AV interface are in communication connection with the main circuit board through interfaces of the 12S channel, respectively.

Preferably, the microphone circuit board is uniformly provided with 6 main microphones along the circumferential direction.

Preferably, the USB interface is a micro-USB interface and/or a C-type USB interface

The utility model provides an external audio input equipment's advantage lies in: the microphones of various different models are arranged in the audio input equipment, so that output signals of different microphones under the same sound source can be acquired, a voice recognition algorithm can be trained based on various different output signals of the same content, the adaptability of the algorithm is improved, and the microphone sensors of different brands and models can be adapted.

Drawings

Fig. 1 is an internal structure diagram of an external audio input device according to an embodiment of the present invention;

fig. 2 is a schematic diagram of a housing of an external audio input device according to an embodiment of the present invention;

fig. 3 is an audio processing flow of the external audio input device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings.

The embodiment provides an external audio input device, which comprises a plurality of microphones with different models; according to the embodiment, the microphones of various different models are arranged in the audio input equipment, so that the output signals of different microphones under the same sound source can be acquired, the speech recognition algorithm can be trained based on various different output signals of the same content, the algorithm adaptability is improved, and the microphone sensors of different brands and models can be adapted.

With reference to fig. 1 and fig. 2, the audio input device further includes a housing 1, and a main circuit board 3, a USB interface 5 and a microphone circuit board 7 communicatively connected to the main circuit board 3, and an AV interface 6 communicatively connected to the main circuit board 3 are disposed inside the housing 1; the microphone is fixed on the microphone circuit board 7 by soldering, and the sockets of the USB interface 5 and the AV interface 6 are respectively positioned on the surface of the shell 1. In the embodiment, the AV interface 6 is used for extracting the reference sound signal to perform echo cancellation, the USB interface 5 is used for receiving external power supply and outputting sound signals picked up by microphones of different models and the reference sound signal extracted through the AV interface 6 to the outside, and since the signal extracted by the AV interface 6 is used for optimizing the microphone signal, the AV interface 6 can also be directly connected with the microphone circuit board 7 in a communication manner. The electronic equipment is connected with the audio input equipment through the USB interface 5 to carry out power supply and signal communication; the USB interface 5 can select a micro-USB interface or a C-type USB interface as required, or both the two USB interfaces are configured. The USB interface is a very common interface for consumer electronics, can be adapted to different intelligent electronic equipment by adopting the USB interface, and has strong adaptability, convenient use and stable performance; the AV interface is adopted to recover the reference sound signal, so that the requirements for realizing different speech processing algorithms can be met, different operating systems can be effectively adapted to carry out secondary development, and the function of the built-in algorithm of the electronic equipment is effectively exerted.

In the preferred embodiment, the auxiliary microphones of different models can be fixed by arranging the auxiliary microphone circuit board 9, the main microphones 8 are arranged on the microphone circuit board 7 in a circumferential array, the AV interface 6 is also in communication connection with the auxiliary microphone circuit board 9, the main microphones 8 arranged on the microphone circuit board 7 in an array collect high-quality audio signals, and the auxiliary microphones of different models are arranged on the auxiliary microphone circuit board 9 to collect respective sound signals, so that the high-quality audio signals collected by the main microphones 8 are used as a standard to identify the signals collected by the auxiliary microphones during training, and an algorithm is trained. And the main microphone and the auxiliary microphone mainly used for training are respectively fixed on different circuit boards, so that the aim of picking up sound signals by using microphones of different models can be fulfilled without changing the structure of the existing audio input equipment.

Specifically, 6 main microphones 8 are uniformly arranged on the microphone circuit board 7 along the circumferential direction, the annular microphone array has strong directivity, and only sound sources in the directional range can be picked up, so that interference noise in different environments can be effectively shielded; the orientation range of the microphones of the circular array is 0-360 degrees, the array topological structure is wide in orientation range and high in sound source positioning precision, the requirements of various voice signal processing algorithms such as sound source positioning and the like can be met, and user experience is good; of course, a person skilled in the art can also adaptively select a linear array of microphones with an orientation range of 0-180 ° for sound pickup.

The position relation of each part in the shell 1 is not strictly limited, the microphone circuit board 7 and the auxiliary microphone circuit board 9 are preferably and directly fixed with the shell in a fixing mode, the microphone circuit board 7 and the auxiliary microphone circuit board 9 are fixed in the shell 1 in parallel, the main circuit board 3 is fixedly matched with the shell 1 and limited between the microphone circuit board 7 and the auxiliary microphone circuit board 9, the auxiliary microphone circuit board 9 is arranged below the main circuit board 3, and the USB interface 5 and the AV interface 6 are fixed below the auxiliary microphone circuit board 9, so that the internal structure is optimized, and the equipment space is reduced. The main circuit board 3 may also be disposed below the auxiliary microphone circuit board 9, the microphone circuit board 7 may be disposed above the auxiliary microphone circuit board 9, and a lead wire connecting the microphone circuit board 7, the main circuit board 3, and the interface circuit board 4 may be reserved on the auxiliary microphone circuit board 9. It should be noted that fig. 1 does not employ the above-described structural relationship of the preferred embodiment for convenience of showing all the structures.

The preferred embodiment can also connect the interface circuit board 4 on the main circuit board 3 through the pin plate 2, the interface circuit board 4 is separated from the main circuit board 3, and the main circuit board 3 is connected through the pin plate 2, so that the user can conveniently and freely customize the interface circuit board 4, and the cost optimization by the user is facilitated.

Referring to fig. 3, in the present embodiment, the microphone circuit board 7, the auxiliary microphone circuit board 9, and the AV interface 6 are respectively connected to the main circuit board 3 through interfaces of 12S channels; the sound signal picked up by the microphone circuit board 7 and the reference sound signal picked up by the AV interface 6 are transmitted to the main circuit board 3 through the interface of the 12S channel; a plurality of sound signals collected by a microphone on the auxiliary microphone circuit board 9 are connected to the main circuit board 3 through another group of 12S channel interfaces, the main circuit board 3 packages and temporarily stores the signals in a buffer area of the main circuit board according to a data format of multi-channel digital audio, and the upper computer reads and operates the temporarily stored data through a USB interface; therefore, the audio signal is processed through the upper computer and used for training the algorithm.

An IIS interface can be arranged on the main circuit board 3 according to needs, secondary development can be carried out by directly using the IIS hardware interface, the USB interface can still be used for supplying power at the moment, and other power supply modes or an internal power supply can also be used for supplying power.

The external audio input device provided by the embodiment has the following advantages:

1. the sound signals picked up by the microphone fixing plate 7 are used for the main functions of the upper computer, for example, the picked sound signals are enhanced and then are subjected to voice recognition or speaker recognition, and the sound signals picked up by the auxiliary microphone circuit board 9 can be sent to a background server for model training of a voice recognition or speaker recognition algorithm, so that the algorithm has a better adaptability effect on microphones of various brands and models; therefore, the audio acquisition effect of the electronic equipment can be expanded almost without changing the existing hardware structure of the electronic equipment, and the user experience is improved;

2. the implementation cost of the scheme is obviously reduced; the annular microphone array has super-strong directivity, only sound sources in a directional range can pick up sound, and interference noise in different environments can be effectively shielded; the method has the advantages that the orientation range is wide, the sound source positioning precision is high, and algorithms in various speech signal processing fields such as sound source positioning can be efficiently realized; the adaptability is strong, the use is convenient, and the performance is stable; the method can meet the requirements of realizing different voice processing algorithms, can effectively adapt to different operating systems for secondary development, and efficiently exerts the function of the built-in algorithm of the electronic equipment.

3. The interface circuit board 4 is separated from the main circuit board 3, and is connected with the main circuit board 3 through the inserting needle plate 2, so that a user can conveniently and freely customize the interface circuit board 4, and the cost optimization by the user is facilitated.

4. Can avoid using the USB interface according to developer's demand, directly use IIS hardware interface to carry out secondary development, the power supply mode of microphone array can select to remain the USB interface and only do the mains operated use this moment, perhaps selects to pull down the interface circuit board and supplies power through plus supply socket, and the practicality is strong and maneuverability is high, has improved developer's work efficiency.

Claims (6)

1. An external audio input device, comprising: the microphone comprises a plurality of microphones with different models which can respectively pick up the same sound signal; the device comprises a shell, a main circuit board, a USB interface, a microphone circuit board and an AV interface, wherein the USB interface and the microphone circuit board are in communication connection with the main circuit board, the AV interface is used for collecting reference sound signals, and a microphone is welded on the microphone circuit board; the sockets of the USB interface and the AV interface are positioned on the surface of the shell; the AV interface is in communication connection with the microphone circuit board or the main circuit board; still be provided with the vice microphone circuit board with main circuit board communication connection in the casing, the even array welding of edge circumferencial direction has a plurality of main microphones on the microphone circuit board, and the welding has the vice microphone of a plurality of different models on the vice microphone circuit board, the AV interface still with vice microphone circuit board communication connection.

2. The external audio input device of claim 1, wherein: the microphone circuit board and the auxiliary microphone circuit board are parallel to each other and fixed inside the shell, the main circuit board is fixedly matched with the shell and limited between the microphone circuit board and the auxiliary microphone circuit board, the auxiliary microphone circuit board is arranged below the main circuit board, and the USB interface and the AV interface are fixed below the auxiliary microphone circuit board.

3. The external audio input device of claim 1, wherein: the main circuit board is also connected with an interface circuit board through an insertion needle plate.

4. The external audio input device of claim 1, wherein: the microphone circuit board, the auxiliary microphone circuit board and the AV interface are in communication connection with the main circuit board through interfaces of the 12S channel respectively.

5. The external audio input device of claim 1, wherein: and 6 main microphones are uniformly arranged on the microphone circuit board along the circumferential direction.

6. The external audio input device of claim 1, wherein: the USB interface is a micro-USB interface and/or a C-type USB interface.

Images