# US 2023/0054740 A1 · Audio generation method, related apparatus, and storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0054740 A1
- Document ID: US20230054740A1
- Status: Published application
- Kind: A1
- Application number: 17/794,540
- Filed: 2020-10-27
- Published: 2023-02-23
- Earliest claimed: 2020-01-22
- Assignees: Petal Cloud Technology Co., Ltd.
- Applicants: Petal Cloud Technology Co., Ltd.
- Inventors: Nian LI
- Web page: https://patentgrep.com/patent/US20230054740A1

## CPC

- G06F 16/685: {using automatically derived transcript of audio data, e.g. lyrics (speech recognition G10L15/00)}
- G06F 16/632: Query formulation
- G06F 16/638: Presentation of query results
- G10H 1/0008: {Associated control or indicating means}
- G10L 13/06: Elementary speech units used in speech synthesisers; Concatenation rules
- G10H 2220/011: Lyrics displays, e.g. for karaoke applications
- G10H 2240/141: Library retrieval matching, i.e. any of the steps of matching an inputted segment or phrase with musical database contents, e.g. query by humming, singing or playing; the steps may include, e.g. musical analysis of the input, musical feature extraction, query formulation, or details of the retrieval process
- G10H 2250/455: Gensound singing voices, i.e. generation of human voices for musical applications, vocal singing sounds or intelligible words at a desired pitch or with desired vocal effects, e.g. by phoneme synthesis
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10H: ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Embodiments of this application provide an audio generation method, a related apparatus, and a storage medium, to provide a better audio generation solution for a user. In embodiments of this application, a text is obtained, a song clip corresponding to the text is obtained through matching, and the song clip is used as audio corresponding to the text. In this way, the text can be expressed in a manner of the song clip.

## Claims

### Claim 1

An audio generation method, comprising:
obtaining a text; and
in response to a song clip corresponding to the text being obtained through matching, using the song clip as audio corresponding to the text, wherein lyrics of the song clip are the same as the text.

### Claim 2 (depends on 1)

The method according to claim 1, wherein in response to no song clip corresponding to the text being obtained through matching, the method further comprises:
dividing the text into at least two subtexts according to a preset rule, wherein the at least two subtexts comprise a first subtext and a second subtext, wherein the first subtext comprises one or more consecutive characters in the text, and wherein the second subtext comprises one or more consecutive characters in the text except the characters comprised in the first subtext; and
in response to a song clip corresponding to the first subtext and a song clip corresponding to the second subtext are obtained through matching, splicing the song clip corresponding to the first subtext and the song clip corresponding to the second subtext to obtain the audio corresponding to the text.

### Claim 3 (depends on 2)

The method according to claim 2, wherein in response to a character comprised in the first subtext being the same as a character comprised in the second subtext, the song clip corresponding to the first subtext and the song clip corresponding to the second subtext meet any one of the following content:
the song clip corresponding to the first subtext and the song clip corresponding to the second subtext are from two different songs;
the song clip corresponding to the first subtext and the song clip corresponding to the second subtext are two different song clips of a same song; and
the song clip corresponding to the first subtext and the song clip corresponding to the second subtext are a same song clip of the same song.

### Claim 4 (depends on 2)

The method according to claim 2, wherein the dividing the text into at least two subtexts according to a preset rule comprises:
allocating remaining characters in the text except the last character as the first subtext, and allocating the last character in the text as the second subtext.

### Claim 5 (depends on 2)

The method according to claim 2, wherein in response to no song clip corresponding to the first subtext being obtained through matching, the method further comprises:
allocating remaining characters in the text except the last two characters as a third subtext, and allocating the last two characters in the text as a fourth subtext; and
in response to a song clip corresponding to the third subtext and a song clip corresponding to the fourth subtext are obtained through matching, splicing the song clip corresponding to the third subtext and the song clip corresponding to the fourth subtext to obtain the audio corresponding to the text.

### Claim 6 (depends on 2)

The method according to claim 2 wherein in response to the song clip corresponding to the first subtext being obtained through matching, but no song clip corresponding to the second subtext being obtained through matching, the method further comprises:
dividing the second subtext into at least two subtexts, wherein the at least two subtexts comprise a fifth subtext and a sixth subtext;
obtaining, through matching, a song clip corresponding to the fifth subtext, and
obtaining, through matching, a song clip corresponding to the sixth subtext; and
splicing the song clip corresponding to the first subtext, the song clip corresponding to the fifth subtext, and the song clip corresponding to the sixth subtext to obtain the audio corresponding to the text.

### Claim 7 (depends on 1)

The method according to claim 1 wherein the obtaining, through matching, a song clip corresponding to the text comprises:
obtaining, through matching, one or more songs corresponding to the text, wherein lyrics of the song comprise the text;
in response to there being a plurality of songs:
selecting one song from the plurality of songs as a target song according to the preset rule or selecting one song from the plurality of songs as a target song based on user selection; and
capturing, from the target song, the song clip corresponding to the text.

### Claim 8 (depends on 7)

The method according to claim 7, wherein the selecting one song from the plurality of songs as a target song according to the preset rule comprises:
selecting one song from among the plurality of songs as the target song based on one or more of the following contents of each of the plurality of songs:
playback times, a singer, and a style of the song.

### Claim 9 (depends on 1)

The method according to claim 1 wherein the obtaining a text comprises:
obtaining a paragraph input by a user, wherein the paragraph includes at least one punctuation mark therein; and
dividing the paragraph into at least two texts according to the punctuation mark included in the paragraph.

### Claim 10 (depends on 1)

A server, wherein the server comprises a processor, a memory, and a communication interface, the memory comprises program instructions, and when the program instructions are executed by the processor, the server is enabled to perform the method according to claim 1.

### Claim 11 (depends on 1)

A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer-executable instructions; and when the computer-executable instructions are invoked by a computer, the computer is enabled to perform the method according to claim 1.