TWI231470B - Automatic learning method for voice parameter - Google Patents

Abstract

The present invention provides an automatic learning method for voice parameters, which is applied on a device with a data access arrangement device interface. The learning method comprises first to abstract different communication voices for analysis during voice communication first time by a user using the device; then, learning the voice parameters and storing the parameters; so as to accurately determine the current status based on different voice parameters and proceed the corresponding operation to prevent the occurrence of faulty operation. The present invention can avoid the trouble of obtaining and inputting the voice parameters, and achieve the effect of simple and convenient use.

Description

1231470

V. Description of the Invention u) Field of the Invention The present invention relates to a technology for learning sound parameters, in particular to a method for automatically learning and distinguishing sound parameters, which is applied to VoIP (Voice over Internet Protocol, Internet) Voice protocol) gateways, automatic voice response systems, switches, and other devices with DAA (Data Access Arrangement) interfaces. Background of the invention: Generally speaking, when conducting a voice communication, the user usually uses the telephone to exchange telecommunication networks through the public.

(Network, PSTN] Swi tched) The disconnection sent by the PSTN should be hung up, the corresponding action, but with the technology equipment, such as the network used for audio communication, the default PSTN is used to test the PSTN. The action given. However, the frequency and beat parameters are preliminarily used by a special instrument commonly known as the telecommunications bureau for circuit connection (Circuit is used to make a call, so the user can determine whether there is a connection or a miss according to the same sound signal. It can be easily judged without any difficulty. The industry is booming. At present, many DAA interfaces are used to dial = Internet (1 P) phones for Internet language or general automatic voice response systems. It is = 1 parameter to judge the current state by itself. It must first determine the current state and make a corresponding response. Two f must know in advance that all sounds (Tone) can correctly detect sounds; User mouth summer test or * pSTN system manufacturers provide all sound

1231470

V. Explanation of the invention (2) The parameters are not identical but this' method will cause loss. Therefore, the present invention is to learn the sound parameters and overcome the conventional shortcomings. Purpose and summary of the invention. The main purpose of the invention. 5 It is another method of the present invention that allows the user to use it easily and conveniently. It can increase the detection and judgment of the current state to make a contrived situation. 0 The method of the present invention enables the voice parameters between communication systems to determine whether the question is 0. In order to achieve the above η step: first confirm that the line phone is connected to the public exchange for the sound parameters sent by the telecommunications network, and finally put forward a problem of automatic learning of sound parameters to obtain and input sound parameters. An accuracy of an automatic learning method of sound parameters is proposed, and actions corresponding to the sound parameters are correctly matched to effectively prevent misoperation. A parameter setting of a sound language that can be learned at any time is provided correctly, so as to avoid disconnection according to the difference or should be avoided. The question of generating a bite line The learning method of the present invention includes the following prepared steps, if so, dial the phone and the road; then the public voice will be obtained, analyze the communication voice and learn whether there are other communication voices. Need to learn

Page 5 1231470

V. Description of the invention (3) Xi, if 疋 then repeat the voice of the general after learning; as: j = j learning. The present invention refers to the sound and learns the current state energy number. According to the correct judgment of “1”, the corresponding action can be made by Yu_Chenxin. It is easy to understand the embodiments of the present invention and the accompanying drawings for detailed explanation. The technical content, characteristics and its achievement: the description of the work figure number: 10 gateway 12 digital signal processing 14 data access arrangement 16 port 18 public switched telecommunication device (DAA) interface network (PSTN ) It is used in switches and other network gate tone responses. It uses the gate to automatically learn the corresponding sound parameters and communicates with the phone in detail. The present invention responds to the system and uses the device to make calls and automatic messages. It can be used for the first time, and it should be used to stop the bite line in a timely manner. The general network gateway has a data storage device, which can be used in the system, switch router or automatic learning of telecommunication networks. It also provides a messenger. Borrow underneath (Gateway: Get the device to call the Internet and call the general electric voice response system to determine the reason why the communication is simpler to use. There is a software in the voice communication system by a network gate), automatic voice (DAA) interface (IP) telephone, and the voice of the voice refers to the communication voice. It is convenient and effective to use the network channel through the public $

Page 6 1231470 V. Description of the invention (4) The embodiment of the network for voice communication by a person changing electricity to explain the function of the present invention. The third figure is a block diagram of a network gateway used in the present invention. The figure ^ has a digital signal processor in the gateway 1 ().

Digital Signal Processor (Dsp) 12, which uses a daA) surface 14 to connect to a port 丨 6, to provide the connection to the general public traffic network (psTN) 18, users use the idle channel for the first time When the device is 10, the digital signal processor 12 is used to detect the communication sound sent from the PSTN 18 and automatically store the frequency and cadence of the communication sound for comparison. Sound parameters. Then, in conjunction with the work flow chart of the second figure, the work flow of automatic learning of sound parameters will be invented in detail. As shown in the figure, the entire satellite work process will not start in step S10, and then the step S12 will check whether the line is ready. If it is not ready, repeat this step. If it is ready, go to step s 1 4. After confirming that the line is ready, proceed to step ^ 4, the phone directly dials the phone number or enters the preset number of the network closed circuit. Make a call ... through; road = public switched telecommunications network 18 to contact the callee on the other end to communicate. If the user dials in step S14, no matter what the communication is on the phone, PSTN 1 8 will respond to the corresponding communication voice. The heart is shown in step S16, which makes the digital signal processor 12 The inner two soft two can be measured t = the communication sound sent by the PSTN 18, and this communication sound can be obtained = fetch ° ° Seoul "for dialing

Page 7 1231470

Dial tone, busy tone, alerting tone, ring back tone, or on-hook tone. Then, as shown in step s 18, after the digital signal processor 丨 2 obtains its communication sound, since each communication sound has its own regular frequency and beat, it must first analyze the cheek of the communication sound. f is used as the sound parameter for comparison, so the digital signal processor 19 1 stands: f communication sounds are analyzed, and the frequency and beat 3 f sounds are learned to learn the current state, and then the corresponding actions are automatically made such as Action to cut off voice communication, etc. In the end, if it is judged in step 32, if there are other communication sounds that need to be related to: If there is no other communication sounds that need to be learned after the break, it will be as in step s22 to 22; if there are other communication sounds that need to be learned, return to At step s 1 2, the above steps are repeated until the end of the learning. Under the condition of more hardware structure, fmeming uses the software inside the device with daa interface such as network gateway 2 automatic voice response system to automatically detect and learn the sound parameters of each communication sound. ΐ: Eliminate the trouble of obtaining and inputting sound parameters. 'People who achieve the convenience of early use and operation can also increase the accuracy of the detection sound, so that the sound parameters of the communication system are correctly set, and according to different sound parameters. The current state (whether it is turned on or should be disconnected) is made to correspond to the current situation. This can effectively prevent the occurrence of misoperation and avoid the occurrence of bite.

Page 8 1231470 V. Description of the Invention (6) Problem. The above-mentioned embodiments are only for explaining the technical ideas and characteristics of the present invention. The purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. When the scope of the patent of the present invention cannot be limited, That is, any equivalent changes or modifications made in accordance with the spirit disclosed in the present invention should still be covered by the patent scope of the present invention.

Page 9 1231470 Brief description of the drawings Description of the drawings: The first diagram is a block diagram of the gateway used in the present invention. The second figure is a schematic diagram of the work flow of the present invention. Page 10

Claims (1)

1231470 ^^^ ^ An automatic learning method of sound parameters, which is applied to a device having a data access = row device (DAA) interface to perform I ° through the device. The method includes the following steps: (A) Use the device to make a call to connect to the public switched telecommunications network (pSTN); (β) capture different communication sounds sent by the PSTN; (c) analyze the communication sound to learn its sound parameters Travel Reserve 6. The scope of patent application is reached, and (D) judge whether there are other communication voices to learn, if yes, go back to step (A), if not, end the learning. 2 · The automatic learning method of sound parameters as described in item 1 of the patent application range, wherein the device is selected from a gateway, an automatic voice response system, and the like. 、, 3 · Automatic learning of sound parameters as described in item 1 of the scope of patent application. Fang Ru which includes a step to check whether the line is accurate before performing step (A), if not, repeat this step, and if it is, proceed to step (A). 4. The sound parameter automatic learning method as described in item 丨 of the scope of the patent application, wherein the communication sound is selected from the group consisting of a dial tone (d 丨 a 丨 tone), a busy tone, a warning tone (aierting tone), A group consisting of ring back tone, disconnection tone (On-h00k tone), etc. Please use the sound parameters described in item 1 of the automatic range, and learn the method, i sound ί t sound "is a regular frequency and beat 'to learn it as 6. sound as described in item 1 of the scope of patent application Automatic parameter learning method, Page 11 1231470 6. Scope of patent application The device uses a digital signal processor connected to the DAA interface to learn and analyze the communication sound. 7. A method for automatic learning of sound parameters, which is applied to a device having a data access arrangement (DAA) interface for voice communication through the device. The method includes the following steps: (A) Use the device to make a call To connect to a telecommunications network; (B) capture different communication sounds sent by the telecommunications network; (C) analyze the communication sound to learn its sound parameters and store it; and (D) Determine if there are other communication sounds to learn, if yes, go back to step (A), if not, end the learning. 8. The voice parameter automatic learning method as described in item 7 of the scope of patent application, wherein the device is selected from a gateway, an automatic voice response system, and a switch. 9. The method for automatic learning of sound parameters as described in item 7 of the scope of patent application, which includes a step of checking whether the line is ready before performing step (A), if not, repeat this step, and if it is, perform step (A ). 10. The method for automatically learning a sound parameter according to item 7 of the scope of patent application, wherein the communication sound is selected from a dial tone (dia 1 t ο ne), a busy tone, and an alerting tone. , Ring back tone, on-hook tone, etc. 11. The sound parameter automatic learning method as described in item 7 of the scope of the patent application, wherein the communication sound has a regular frequency and tempo to learn it as Page 12 1231470 Page 13