Advertisement voice intelligent generation system and method based on deep learning, and storage medium

Abstract

The invention relates to the technical field of advertisement design, in particular to an intelligent advertisement voice generation system and method based on deep learning and a storage medium, the system comprises a server, and the server comprises the following modules: a demand acquisition module used for acquiring an advertisement dubbing demand of a user and a corresponding dubbing preparation file; the advertisement analysis module is used for carrying out content analysis on the advertisement sample; the voice generation module is used for screening a voice generation model from a voice generation model library according to application scene analysis, and importing the dubbing text into the voice generation model to obtain an advertisement voice initial edition file; the matched music selection module is used for selecting matched background music according to the application scene analysis and the time length of the advertisement voice first edition file; and the sound processing module is used for synthesizing the advertisement voice initial edition file and the background music to obtain an advertisement voice final edition file. According to the invention, the time cost and labor cost of advertisement voice production can be reduced, and the advertising effect of the advertisement is ensured.

Description

technical field [0001] The present invention relates to the technical field of advertisement design, in particular to a system, method and storage medium for intelligently generating advertisement voice based on deep learning. Background technique [0002] The main function of advertisements is to guide consumption, and the visual and auditory impressions created by advertisements will arouse consumers' realistic desire to buy. The current advertising voice is basically dubbed manually, which is troublesome for small and micro businesses and individual merchants. Either the boss or employees will dub in person, or professional voice actors will be hired to dub manually. These methods require frequent auditions and dubbing. It will cost a lot of time and manpower. Secondly, the audio of the current intelligent speech synthesis is relatively flat and boring, and it cannot be used directly in advertisements to achieve a good publicity effect. Contents of the invention [00...

AI Technical Summary

Problems solved by technology

The current advertising voice is basically dubbed manually, which is troublesome for small and micro businesses and individual merchants. Either the boss or employees will dub in person, or professional voice actors will be hired to dub manually. These methods require frequent auditions and dubbing. It will cost a lot of time and manpower

Secondly, the audio of the current intelligent speech synthesis is relatively flat and boring, and it cannot be used directly in the advertisement to achieve the promotional effect of the advertisement.

Images