Query by Humming Ringtone Search via Acoustic Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users find it inconvenient to search for ringtones on mobile devices due to the need for text input, especially when they remember melodies rather than song titles or artists.
Innovation Solution
A technology that allows users to input a rendition of a ringtone, either vocally or instrumentally, which is then processed to match against actual ringtones, using features like pitch contours, intervals, and durations, and returned as a selection set, potentially weighted by popularity and importance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If text-based search methods are used to find ringtones, then search accuracy can be maintained, but user convenience deteriorates due to the need for text input on small devices
Solution Approach 1:
The patent replaces the mechanical text input system with an acoustic input system. Users hum or sing the melody they want to search for, and the system processes this audio input to identify matching ringtones. This substitution eliminates the need for manual text entry on small devices while maintaining search accuracy through audio pattern recognition and matching algorithms.
Solution Approach 2:
The patent introduces an audio processing intermediary system that acts as a mediator between the user's humming input and the ringtone database. This intermediary captures the acoustic signal, extracts melodic features, compares them against stored ringtone patterns, and returns matching results. This intermediary layer enables natural voice-based interaction while preserving precise search capabilities.
2Measurement precision
If users are required to input exact song titles or artists, then search precision is improved, but ease of operation deteriorates since users often cannot recall exact information
Solution Approach 1:
The patent replaces the requirement for precise text recall with an acoustic input method. Users simply hum or sing the melody without needing to remember exact titles or artists. The system's audio processing algorithms extract melodic contours and patterns from the humming, automatically matching them against the ringtone database to achieve precise identification without demanding exact recall from the user.
Solution Approach 2:
The patent changes the search parameter from text-based exact matching to audio-based pattern matching. Instead of requiring users to input precise textual information, the system accepts variable acoustic representations of the same melody and uses signal processing to identify the underlying pattern. This parameter change allows flexible, imprecise user input to yield precise search results.
3Adaptability or versatility
If audio processing is performed to match user input with ringtones, then search capability is enhanced, but device complexity increases
Solution Approach 1:
The patent extracts the complex audio processing functions from the mobile device itself and relocates them to a remote server or cloud-based processing system. The mobile device only needs to capture the humming audio and transmit it to the server, which performs the sophisticated signal processing, pattern recognition, and database matching. This extraction reduces the processing complexity requirements of the mobile device while maintaining enhanced search capabilities.
Data Source
AI summary
Described is a technology by which a user hums, sings or otherwise plays a user-provided rendition of a ringtone (or ringback tone) through a mobile telephone to a ringtone search service (e.g., a WAP, interactive voice response or SMS-based search platform). The service matches features of the user's rendition against features of actual ringtones to determine one or more matching candidate ringtones for downloading. Features may include pitch contours (up or down), pitch intervals and durations of notes. Matching candidates may be ranked based on the determined similarity, possibly in conjunction with weighting criterion such as the popularity of the ringtone and/or the importance of the matched part. The candidate set may be augmented with other ringtones independent of the matching, such as the most popular ones downloaded by other users, ringtones from similar artists, and so forth.