Voice Playlist Generation via Key Tag Extraction and Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice interaction-based multimedia resource playing systems require user intervention for playlist editing, which is inefficient and limits the automation of voice service interactions.
Innovation Solution
A voice interaction method and apparatus that acquires voice request information, identifies key tags for multimedia resources, and generates a playlist based on popularity data and user descriptors, eliminating the need for user editing by automatically selecting and ordering multimedia resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If user manually edits playlist on traditional platform, then user can customize playlist, but operation efficiency is low and automation is limited
Solution Approach 1:
The system automatically generates playlists by extracting voice request information and matching it with multimedia resources from the library, eliminating the need for users to manually edit playlists. The voice interaction system serves itself by autonomously selecting and ordering resources based on user voice commands.
Solution Approach 2:
The patent replaces manual mechanical editing operations with automated voice-based information processing. The system uses voice recognition, natural language processing, and machine learning algorithms to automatically extract key tags, search for matching resources, and generate playlists, substituting manual user actions with automated computational processes.
2Measurement precision
If voice interaction system searches through entire multimedia library, then resource matching accuracy improves, but processing time increases
Solution Approach 1:
The patent segments the multimedia library by extracting key tags from voice requests and using them as search criteria. Instead of searching the entire library, the system divides the search space by identifying specific characteristic attributes (key tags) and filtering resources based on these tags, which reduces search scope while maintaining accuracy.
Solution Approach 2:
The system performs preliminary processing by pre-extracting key tags from voice requests and pre-organizing multimedia resources with their corresponding tags in the library. This preliminary action enables faster matching during actual playlist generation, reducing processing time while maintaining high resource matching accuracy.
Data Source
AI summary
Embodiments of this disclosure disclose a voice interaction based method and apparatus for generating a multimedia playlist. An embodiment of the method comprises: acquiring first voice request information for playing multimedia resources; identifying a key tag for indicating a characteristic attribute of the multimedia resources in the first voice request information; finding the multimedia resources having the key tag in a multimedia resource library; and generating a multimedia playlist based on the found multimedia resources. The embodiment realizes automatic generation of multimedia playlists and improves the efficiency of voice service.