Search

CN-122022796-A - Voice pricing method and related products

CN122022796ACN 122022796 ACN122022796 ACN 122022796ACN-122022796-A

Abstract

The invention provides a voice pricing method and related products, and relates to the technical field of commodity pricing. The voice pricing method comprises the steps of obtaining preset voice information V a =｛v 1 ,v 2 ,v 3 ,...,v n , wherein V 1 ,v 2 ,v 3 ,...,v n is voice information corresponding to n commodity information respectively, obtaining real-time voice information V b , finding out voice information V t with highest acoustic feature similarity with the real-time voice information V b in the preset voice information V a , matching commodity information corresponding to the voice information V t with the real-time voice information V b , and pricing according to pricing rules. Related products include computer-readable storage media, computer devices, computer program products, voice pricing devices, and voice pricing electronic scales. The invention can enable the user to use voice to pay and settle accounts for the commodity according to own language habits, thereby enabling the commodity pricing process to be more efficient and convenient.

Inventors

  • WANG WANDONG

Assignees

  • 王宛东

Dates

Publication Date
20260512
Application Date
20230918

Claims (10)

  1. 1. A method of voice pricing, comprising: acquiring preset voice messages V a =｛v 1 ,v 2 ,v 3 ,...,v n , wherein V 1 ,v 2 ,v 3 ,...,v n is voice information corresponding to n commodity information respectively; Acquiring real-time voice information V b ; Finding out the voice information V t with highest acoustic feature similarity with the real-time voice information V b from the preset voice information V a , and matching commodity information corresponding to the voice information V t with the real-time voice information V b ; And pricing is carried out according to pricing rules.
  2. 2. The voice pricing method according to claim 1, wherein V 1 ,v 2 ,v 3 ,...,v n is voice information corresponding to n pieces of commodity information, the commodity information is recorded in advance by a user, the preset voice information V a is recorded by the user according to pronunciation habits of individuals, and the voice information comprises voice intensity and a waveform chart with time as a horizontal axis and amplitude as a vertical axis.
  3. 3. The voice pricing method according to claim 1, wherein the voice information of the commodity information to be billed sent in real time by the user in the commodity transaction pricing process of the real-time voice information V b Is that is obtained in the real-time voice information V b , and the voice information comprises voice intensity and a waveform chart with time as a horizontal axis and amplitude as a vertical axis.
  4. 4. The voice pricing method according to claim 1, wherein the step of finding out the voice information V t with highest similarity to the acoustic characteristics of the real-time voice information V b from the preset voice information V a , and matching the commodity information corresponding to the voice information V t with the real-time voice information V b comprises the steps of: The method comprises the steps of carrying out framing, windowing and noise reduction on waveform diagram data W a =｛w 1 ,w 2 ,w 3 ,...,w n corresponding to preset voice information V a , wherein W 1 ,w 2 ,w 3 ,...,w n is waveform diagram data of voice information of n commodity information respectively, converting each processed waveform diagram into a spectrogram sequence set L a =｛l 1 ,l 2 ,l 3 ,...,l n respectively, wherein L 1 ,l 2 ,l 3 ,...,l n is a spectrogram sequence corresponding to the n commodity information respectively, converting the spectrogram sequence into a three-dimensional spectrogram sequence set S a =｛s 1 ,s 2 ,s 3 ,...,s n respectively, wherein S 1 ,s 2 ,s 3 ,...,s n is a spectrogram sequence corresponding to the n commodity information respectively, extracting characteristic values of each spectrogram sequence in the three-dimensional spectrogram sequence set S a , calculating hash codes H a =｛h 1 ,h 2 ,h 3 ,...,h n of voice information characteristic data of the obtained commodity information, wherein H 1 ,h 2 ,h 3 ,...,h n is hash codes corresponding to the n commodity information respectively, and recording the obtained hash codes H a ; The method comprises the steps of performing framing, windowing and noise reduction on a waveform diagram W b of real-time voice information V b of commodity information to be charged, converting the waveform diagram into a spectrogram sequence L b , converting the spectrogram sequence into a three-dimensional spectrogram sequence S b of the commodity information to be charged, extracting characteristic values on the three-dimensional spectrogram sequence S b of the commodity information to be charged, and calculating hash codes H b of the commodity information to be charged; The method comprises the steps of respectively calculating Similarity = { sim 1 ,sim 2 ,sim 3 ,...,sim n }, wherein sim 1 ,sim 2 ,sim 3 ,...,sim n is the Similarity between commodity information to be charged and stored n commodity information, by using hash codes H b of commodity information to be charged and hash codes H a of stored n commodity information; And comparing the maximum sim t in sim 1 ,sim 2 ,sim 3 ,...,sim n , and taking commodity information corresponding to the voice information v t as commodity information to be priced.
  5. 5. A computer readable storage medium, characterized in that a computer program is stored on the computer readable storage medium, which computer program, when being executed by a processor, carries out the steps of the method according to any one of claims 1-4.
  6. 6. A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of the method according to any one of claims 1-4 when the computer program is executed.
  7. 7. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1-4.
  8. 8. The voice pricing device is characterized by comprising an acquisition module, a matching pricing module and an output module, wherein the matching pricing module is respectively connected with the acquisition module and the output module, the acquisition module is used for acquiring voice information, the matching pricing module is used for executing the steps of the method in any one of claims 1-4, and the output module is used for outputting pricing results.
  9. 9. The voice pricing electronic scale is characterized by comprising an acquisition module, a weighing module, a matching pricing module and an output module, wherein the matching pricing module is respectively connected with the acquisition module, the weighing module and the output module, the acquisition module is used for acquiring voice information, the weighing module is used for measuring weight information of commodities, the matching pricing module is used for acquiring the weight information of the commodities through the weighing module and executing the steps of any one of claims 1-4, and the output module is used for outputting pricing results.
  10. 10. The voice-pricing electronic scale of claim 9, wherein the matching pricing module is further configured to perform the steps of: If the total weight of the commodity in the P (P > 0) time is x and the total weight of the commodity in the (P+1) time is y, the weight of the commodity added in the (P+1) time is set as (y-x).

Description

Voice pricing method and related products Technical Field The invention relates to the technical field of commodity pricing, in particular to a voice pricing method and related products. Background In the commodity transaction process, the commodity can be settled after being priced. In the related devices for pricing by analyzing the voice of the user, a voice recognition system is generally used to process the voice sent by the user, then a language model is used to convert the voice into text, and then the price is matched with the price of the corresponding commodity. However, in practical application, the problem is that on one hand, the pronunciation of users with different language habits (usually called as 'north-south accent') does not always meet the standard, and on the other hand, the accuracy of voice recognition is not high because some people usually speak and vomit words or stuttering, and on the other hand, the calling modes of people with different regions on the same commodity are different, so that the users feel very awkward to use. For example, tomatoes, there are many different names "tomatoes", "foreign persimmons", "solanaceous fruits", "red persimmons" and so on in different places of our country. If a user refers to a "tomato" as "fan shi" at ordinary times, it is difficult to match the price of a tomato for pricing when selling the tomato even if the voice recognition system recognizes the text "persimmon". It is very difficult for many users to use a uniform name and speak standard mandarin for users with different language habits or pronunciation patterns during transactions. The inventor believes that this is why voice recognition technology is currently difficult to apply independently in transaction pricing. Although the current voice recognition pricing is theoretically feasible, the personalized pronunciation requirements of different users are difficult to meet in the actual pricing settlement process, and the voice recognition pricing only has the function of assisting the users to confirm commodities, is difficult to completely replace manual input, code scanning settlement or image recognition settlement, and is difficult to be widely accepted by the users. Disclosure of Invention In the retail trade of goods, although the communication between the buyer and the seller may be different, the information of the goods related to the price is very limited, mainly including the name of the goods, the characteristics of the goods (such as brand, specification, model, place of production, quality classification), the amount of the trade (quantity or weight), the unit price, the discount mode, the price order, etc. The commodity to be priced can be priced only by distinguishing the voice information of the commodity information to be priced. In order to solve the problems in the background technology, the invention has the thought that a user speaks the voice information of commodity information according to own language habit and records in advance, when in transaction pricing, the real-time voice information sent by the user is compared with the voice information recorded in advance, the item with the highest similarity with the acoustic characteristic of the item is found out from the recorded voice information, and the commodity to be billed can be matched, and finally the pricing is completed. Therefore, the voice pricing method and the related products provided by the invention aim to enable commodity pricing process to be more convenient and efficient. In a first aspect of the present invention, a method for voice pricing includes: S1, acquiring preset voice messages V a=｛v1,v2,v3,...,vn, wherein V 1,v2,v3,...,vn is voice information corresponding to n pieces of commodity information respectively; S2, acquiring real-time voice information V b; S3, finding out voice information V t with highest acoustic feature similarity with the real-time voice information V b from preset voice information V a, and matching commodity information corresponding to the voice information V t with the real-time voice information V b; S4, pricing is carried out according to pricing rules. In step S1, the commodity information is recorded in advance by a user, the commodity information comprises commodity names, commodity characteristics, price units, unit price, transaction amount and price instructions, the preset voice information V a Is that is formed by the fact that the user pronounces and records the commodity information according to the pronunciation habits of the user, and the voice information comprises voice intensity and a waveform chart taking time as a horizontal axis and amplitude as a vertical axis. Further, in step S2, the real-time voice information V b refers to voice information of the commodity information to be charged, which is sent by the user in real time during the commodity transaction charging process, and the voice information includes the sound intensity of vo