CN213716506U

CN213716506U - Echo cancellation device

Info

Publication number: CN213716506U
Application number: CN201922435690.4U
Authority: CN
Inventors: 胡佩佩; 高可攀
Original assignee: SHENZHEN GRANDSTREAM NETWORKS Inc
Current assignee: SHENZHEN GRANDSTREAM NETWORKS Inc
Priority date: 2019-12-30
Filing date: 2019-12-30
Publication date: 2021-07-16
Anticipated expiration: 2029-12-30

Abstract

The embodiment of the utility model discloses echo cancellation device, this echo cancellation device mainly includes: the system comprises an audio processing chip, a network module, a microphone and a loudspeaker; wherein, the network module is connected with the audio processing chip; the microphone is connected with the audio processing chip through the analog/digital converter; the loudspeaker is connected with the audio processing chip through a digital-to-analog converter and a digital power amplifier in sequence; the audio processing chip is used for performing algorithm processing on the audio signals collected by the microphone, and further comprises an AES67 processing module, an echo cancellation module and a DSP. The embodiment of the utility model provides an effectively utilized the advantage of the high sampling rate of AES67 protocol standard, guaranteed that the near-end signal that local pickup microphone gathered and echo signal's true degree is high, be favorable to accurately estimating echo intensity, be favorable to getting rid of echo signal's composition effectively from exporting the near-end signal to improve echo effect.

Description

Echo cancellation device

Technical Field

The utility model relates to a speech processing technology field, in particular to echo cancellation technique for among the speech transmission electronic equipment.

Background

At present in the 21 st century, more and more organizations and individuals use a conference system or a live broadcast platform to communicate for a long time, when a far-end talks, sound played by a local loudspeaker is collected by a local microphone and then sent back to the far-end, and in addition, the far-end can hear own echo due to the influence of various delays such as network and data processing, and the echo problem directly influences the voice call quality. Therefore, acoustic echo cancellation plays a very important role in ensuring voice call quality in applications such as communications, teleconferencing, and synchronous classrooms.

The currently common echo cancellation scheme is to take a digital signal before a far-end signal is played through a loudspeaker as a reference signal, take a local pickup microphone acquisition signal as a near-end signal, send the near-end signal and the reference signal to an echo canceller together for echo cancellation, and output a local pickup signal left after the near-end signal is removed of echo signal components.

The effectiveness of echo cancellation is limited in many ways, where speech transmission quality and delay estimation are very critical factors. In order to align the reference signal with the near-end signal in the time domain and effectively and accurately remove the echo signal from the near-end signal, delay estimation and high-quality transmission quality are particularly important, and good voice transmission quality and stable and accurate delay estimation have a decisive role in improving the echo cancellation effect.

AES67 protocol (interoperability protocol of high performance audio streams over network protocols), enables transmission of high quality and low latency (less than 10ms) digital audio signals over high performance networks. Therefore, the utility model provides a combine the echo cancellation device of AES67 protocol standard's high quality audio transmission and low time delay advantage can effectively improve echo cancellation's effect.

Disclosure of Invention

An object of the utility model is to provide an echo cancellation device has combined AES67 protocol standard's high quality audio transmission and low time delay advantage to can effectively improve echo cancellation's effect.

In order to solve the above problem, the embodiment of the utility model provides a technical scheme as follows:

an echo cancellation device mainly comprises an audio processing chip, a network module, a microphone and a loudspeaker; the network module is connected with the audio processing chip; the microphone is connected with the audio processing chip through an analog-to-digital converter; the loudspeaker is connected with the audio processing chip through a digital-to-analog converter and a digital power amplifier in sequence; the audio processing chip is used for performing algorithm processing on the audio signal acquired by the microphone, and further comprises an AES67 processing module, an echo cancellation module and a DSP (digital signal processor).

Preferably, the network module is connected to the audio processing chip through an AES67 network cable interface, and is used for network transmission of audio signals.

Preferably, the echo cancellation module further comprises: the device comprises a voice detector, a delay alignment module, an adaptive filter and a residual echo cancellation module, wherein the delay alignment module is used for compensating the delay difference between the data played by the loudspeaker and the data collected by the microphone.

Preferably, the audio processing chip is connected with the power supply assembly through a power interface.

Preferably, the microphone is a digital microphone, supports the AES67 protocol standard, and supports high sampling rate acquisition.

Preferably, the loudspeaker supports the AES67 protocol standard.

Preferably, the analog-to-digital converter is connected with the audio processing chip through an analog signal interface.

Preferably, the digital power amplifier is connected with the audio processing chip through a digital signal interface.

The utility model provides a technical scheme is through the advantage that has combined AES67 standard, and the synchronization mechanism, any receiving terminal all passes through a public clock on the network, has guaranteed that all flows all sample with the same rate and restore, and all receiving terminals can synchronous playback, and this synchronism and real-time that has just guaranteed the audio stream transmission. Meanwhile, the advantages of the AES67 standard of high sampling rate are combined, and the protocol standard supports three sampling frequencies: 44.1kHz, 48kHz and 96kHz are adopted to ensure that the near-end signals and echo signals collected by the local pickup microphone have high trueness, so that the method is favorable for accurately estimating the echo intensity and effectively removing the components of the echo signals from the output near-end signals.

Furthermore, the utility model provides a technical scheme has combined the advantage of the low time delay of AES67 standard, and lower system delay is favorable to reducing the error of the time delay estimation when echo processing, is favorable to solving the unstable scheduling problem of time delay estimation, and the accuracy of time delay estimation can effectively avoid the remaining echo of conversation in-process.

The utility model provides a technical scheme uses in the VOIP conversation, can effectively improve echo cancellation's accuracy, also can improve the duplex effect in the conversation process to a certain extent.

Furthermore, the utility model provides a technical scheme not only is applicable to the conference system, but wide application improves echo cancellation effect in live platform, broadcasting, long-range classroom education etc. moreover.

In order to make the above and other objects of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a schematic structural diagram of an echo cancellation device according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of an echo cancellation principle according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments refers to the accompanying drawings for illustrating the specific embodiments in which the invention may be practiced.

The application example of the echo cancellation device in the embodiment of the present invention is illustrated by taking a voice conference system as an example, and the electronic products applied by the echo cancellation device here include but are not limited to ordinary phones, VoIP phones, conference systems, handheld mobile communication devices, consumer electronic devices, small computers, and other electronic products having functions of voice acquisition, playing, processing, and network transmission.

The embodiment of the utility model provides a built-in echo cancellation device in a complete conference system, its leading principle is: when the conference system receives a signal transmitted from an opposite terminal, the signal is processed by an algorithm and then played by a loudspeaker after being amplified by a power amplifier of an echo cancellation device, a local microphone collects the signal of the loudspeaker and also collects a near-end signal of environmental noise and speech, and the signal is sent to a far end after being respectively subjected to echo processing by an AES67 audio processing chip and a conventional audio processing module.

Fig. 1 is a schematic structural diagram of an echo cancellation device according to an embodiment of the present invention. As can be seen from the figure, a complete echo cancellation device built in a conference system mainly includes: microphone 101, analog-to-digital converter 102, network module 103, power supply component 104, audio processing chip 105, digital power amplifier 106, digital-to-analog converter 107 and speaker 108. Wherein, one end of the analog/digital converter 102 is connected to the microphone 101, and the other end is connected to the audio processing chip 105; the audio processing chip 105 is further connected with a digital power amplifier 106, a digital/analog converter 107 and a loudspeaker 108 in sequence; the network module 103 and the power supply component 104 are also respectively connected to the audio processing chip 105 through the AES67 network interface 1031 and the power interface 1041 for implementing network signal transmission and chip power supply.

The network module 103 is connected to the audio processing chip 105 for facilitating transmission of audio signals between the conference system and the voice device at the opposite end of the network. The microphone 101 is used for acquiring a local audio signal of the conference system and then converting the local audio signal into a digital signal through the analog/digital converter 102, so as to facilitate subsequent further voice signal processing. The digital power amplifier 106, the digital/analog converter 107, and the speaker 108 are mainly used for outputting the audio signal after the algorithm processing. In addition, in fig. 1, the audio processing chip 105 mainly performs algorithm processing on the audio signal collected by the microphone to realize processing of the voice signal collected by the microphone, and specifically includes: AES67 processing module 1051, echo cancellation module 1052, and DSP 1053. The AES67 processing module 1051 mainly supports the AES67 standard protocol, reduces the system delay, reduces the influence of the delay alignment module, and improves the accuracy of the delay alignment module, thereby improving the echo effect. An echo cancellation module 1052, for processing the echo cancellation to eliminate the echo during the call. The DSP1053 and the digital signal processor are specialized microprocessors that convert analog signals into digital signals, perform a series of processing on the digital signals, and convert the digital signals into analog signals through analog-to-digital conversion.

The echo cancellation module 1052 is composed of a voice detector, a delay alignment module, an adaptive filter and residual echo cancellation, wherein the delay alignment module has a great influence on the performance of the whole system, and is mainly used for compensating the delay difference between the data played by the loudspeaker and the data collected by the microphone. The existing VoIP is all digital, which converts the electrical signal into digital signal, collects a segment of signal which may be 20 ms 40 ms, compresses the signal first and then sends out, so the digital processing result is high efficiency, originally one line can only send one telephone, and currently 10 telephones may be sent, but the efficiency is improved, and simultaneously, the problems of delay and echo are brought. The utility model discloses the scheme has combined AES67 protocol standard's audio frequency processing chip, has reduced system delay, has also reduced the influence that the module was aligned in the time delay, has improved the accuracy that the module was aligned in the time delay to improve echo effect.

The utility model discloses echo cancellation device can operate in speech equipment, and this speech equipment can be any one of mobile terminal, palm computer, panel computer, fixed telephone, VoIP phone, conference system etc..

Besides a conference system, the method can be widely applied to a live broadcast platform, broadcasting, remote classroom education and the like in an application scene so as to improve the echo cancellation effect.

Fig. 2 is a schematic structural diagram of an echo cancellation principle according to an embodiment of the present invention. The following describes embodiments of the present invention in detail with reference to fig. 1 and 2.

The audio processing chip 105 is connected to the power supply component 104 and the network module 103 supporting the AES67 standard protocol. When the conference system is started and then makes a supervision call with a remote device through a network, as shown in fig. 2, a voice message sent by a remote signal plays a voice through a speaker 108, and a near-end signal including near-end noise and near-end speaker voice exists around the conference system. The far-end signal X (n) is taken as a reference signal, the far-end signal is played through a digital power amplifier 106 and a loudspeaker 108, and meanwhile, the near-end has a human speaking T (n) and an environmental noise N (n).

The sound played by the loudspeaker 108 is collected by the local microphone 101 through the physical echo channel h (n), and the actual echo generates a signal formula: y (n) = x (n) × h (n).

The embodiment of the utility model provides an in the local microphone 101 that uses be digital microphone, support the collection of high sampling rate to guaranteed the near-end signal of collection and echo signal's high degree of truth, be favorable to accurately estimating echo intensity. The microphone 101 supports the AES67 protocol standard, and facilitates data processing in cooperation with the AES67 processing block in the audio processing chip 105, so that components of the echo signal can be effectively removed from the output near-end signal. The local microphone 101 collects the near-end signal, which is the sum of the near-end speech t (n), the near-end noise n (n), and the echo signal y (n): y (n) = y (n) + n (n) + t (n).

In the specific echo cancellation processing process, a far-end signal is used as a reference signal, the reference signal is aligned with an echo signal acquired by a local microphone through a voice detector and a delay alignment module, and the echo signal is removed through a self-adaptive filter. When the far-end signal of the reference signal is x (n) and the simulated echo channel is h (n), the estimated echo signal is: l (n) = x (n) × h (n).

Echo signals are removed from near-end signals through a self-adaptive filter, then echo participation is removed through NLP (Nonlinear Programming), and finally signals E (n) which are sent to a far end after echo signals are removed are obtained, namely E (n) formula is as follows: e (n) = y (n) -l (n).

The utility model discloses the scheme has utilized advantages such as AES67 agreement standard's high sampling rate, low time delay, can be effectively from the composition of echo signal in the output near-end signal to improve echo effect. The time delay of echo cancellation process contains transmission delay and space delay, the utility model discloses the scheme combines AES67 protocol standard's low time delay characteristics, has reduced transmission delay greatly to compensate the time delay difference between the data of speaker broadcast and the data that the microphone was gathered, reduced the judgement error of system delay and time delay estimation, improved the accuracy of time delay alignment module.

In summary, although the present invention has been described with reference to the preferred embodiments, the above-described preferred embodiments are not intended to limit the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, so that the scope of the present invention shall be determined by the scope of the appended claims.

Claims

1. An echo cancellation device is characterized in that the echo cancellation device mainly comprises an audio processing chip, a network module, a microphone and a loudspeaker;

the network module is connected with the audio processing chip; the microphone is connected with the audio processing chip through an analog-to-digital converter; the loudspeaker is connected with the audio processing chip through a digital-to-analog converter and a digital power amplifier in sequence;

the audio processing chip is used for carrying out algorithm processing on the audio signals collected by the microphone, and further comprises an AES67 processing module, an echo cancellation module and a DSP.

2. The echo cancellation device of claim 1, wherein the network module is connected to the audio processing chip via an AES67 network cable interface for network transmission of audio signals.

3. The echo cancellation device of claim 2, wherein the echo cancellation module further comprises: the device comprises a voice detector, a delay alignment module, an adaptive filter and a residual echo cancellation module, wherein the delay alignment module is used for compensating the delay difference between the data played by the loudspeaker and the data collected by the microphone.

4. The echo cancellation device of claim 3, wherein the audio processing chip is connected to the power supply module via a power interface.

5. The echo cancellation device of claim 4, wherein the microphone is a digital microphone supporting AES67 protocol standard and high sample rate acquisition.

6. The echo cancellation device of claim 5, wherein the speaker supports the AES67 protocol standard.

7. The echo cancellation device of claim 6, wherein the analog-to-digital converter is coupled to the audio processing chip via an analog signal interface.

8. The echo cancellation device of claim 6, wherein the digital power amplifier is connected to the audio processing chip through a digital signal interface.