A modeling method for generating intelligent Harsh music lyrics

A modeling method and lyric technology, applied in the field of deep neural network modeling, can solve the problem that the Seq2Seq model is easy to generate security responses, etc., and achieve the effects of no time and space restrictions, high generation efficiency, and strong anti-noise ability

Active Publication Date: 2019-05-28
XIAMEN UNIV
View PDF5 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Mou et al. [9] A keyword-based text generation model is proposed to solve the problem that the Seq2Seq model is easy to generate secure replies

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A modeling method for generating intelligent Harsh music lyrics
  • A modeling method for generating intelligent Harsh music lyrics
  • A modeling method for generating intelligent Harsh music lyrics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings.

[0049] see Figure 1~3 , the implementation of the present invention comprises the following steps:

[0050] Step 1: Crawl 100,000 hip-hop lyrics from Netease Cloud Music and complete the data cleaning work. It mainly removes dirty sentences, repetitive sentences and unknown characters in the lyrics to form a hip-hop lyrics corpus. The corpus formed after this operation is figure 2 corpus in .

[0051] Step 2: Use the TF-IDF model to extract the subject of each lyrics, and generate a corpus-based thesaurus.

[0052] Step 3: Calculate the similarity between each sentence lyrics in the corpus and the thesaurus using cosine similarity, and assign some semantically closest lyrics to each topic as the first sentence candidates for the generated results. The operation process of step 2 and step 3 to generate thesaurus and related lyrics is as follows: ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a modeling method for generating intelligent Harsh music lyrics. The method includes: completing data cleaning work from the network easy cloud music to form a Harlyric corpus;generating a topic word library based on a corpus; determining lyrics with semantics similar to that of each topic to serve as first sentence candidates of a generation result; extracting pinyin of last 1-5 characters of each sentence of lyric by using initial consonants and vowels of a modern pinyin system; carrying out word segmentation on the lyric data by using an accurate mode of a ghost word segmentation device, and extracting key words of each sentence of lyric; inputting the lyrics into a Word2Vec model to obtain word vectors corresponding to the lyrics; using the obtained word vectors as training data, and using the training neural network model; outputting the rest lyrics by the neural network through a client interface of the webpage; outputting lyrics output by the neural network to an interface of a webpage program; outputting lyrics output by the neural network to an interface of a webpage program by utilizing a client interface of the webpage; and generating corresponding rhyming words and a next sentence of lyrics for reference of a creator.

Description

technical field [0001] The invention relates to a deep neural network modeling method, in particular to a modeling method for intelligent hip-hop music lyrics generation. Background technique [0002] stuttering participle: [0003] Stuttering word segmentation is a kind of word segmentation technology often used in text analysis. It mainly supports three modes, namely precise model, full mode and search engine model. Among them, the precise mode tries to divide the sentence most accurately, which is suitable for text analysis; the full mode can quickly scan all the words that can be formed into words in the sentence. Although this mode is very fast, it cannot solve the ambiguity problem in Chinese ;The search engine mode is to segment long words again on the basis of the precise mode, so as to achieve the purpose of improving the recall rate. This mode is more suitable for search engine word segmentation. [0004] word2vec model: [0005] A word is the smallest independe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/27G06F16/36G06F16/9535
Inventor 孙蒙新刘昆宏王备战洪清启张海英
Owner XIAMEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap