Machine-Learning Audio Blending for Smooth FM Signal Switching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional FM radio receivers switch between FM signals based solely on power levels, leading to deteriorated audio quality and interruptions before the power level drops below the threshold, affecting the listening experience.

Innovation Solution

A system utilizing a digital signal processing (DSP) circuit with a trained machine learning model to analyze audio quality scores of multiple FM signals, blending them based on quality rather than power levels to predict and smooth transitions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If switching between FM signals is based solely on power levels, then the switching decision is simple and reliable, but audio quality deteriorates and interruptions occur before power levels drop below threshold

Engineering Contradiction:
Improveswitching reliabilityVSAvoidoperation simplicity
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent changes the parameter used for switching decisions from power level alone to a combination of power level and audio quality metrics. The DSP circuit monitors both power strength and audio quality parameters (such as signal-to-noise ratio, distortion levels, and other quality indicators) to make informed switching decisions. This allows the system to detect audio quality deterioration early and switch before interruptions occur, resolving the contradiction between reliable switching and operational simplicity.

Inventive Principle:
Principle #35Parameter changes

2Loss of energy

If switching is delayed until power level drops below threshold, then power utilization is maximized, but audio quality has already deteriorated causing interruptions

Engineering Contradiction:
Improvepower utilization efficiencyVSAvoidaudio quality continuity
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The patent implements preliminary action by monitoring audio quality parameters in advance of power level degradation. The DSP circuit continuously evaluates audio quality metrics and predicts potential deterioration trends. When quality metrics indicate impending degradation (even while power levels remain above threshold), the system proactively initiates switching to the alternative FM signal. This preliminary detection and action prevents audio quality deterioration and interruptions while still maximizing power utilization by switching only when necessary.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If continuous monitoring of audio quality parameters is implemented, then audio quality is maintained, but computational complexity and processing requirements increase

Engineering Contradiction:
Improveaudio quality maintenanceVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces complex continuous computational monitoring with a more efficient parameter-based detection system. Instead of continuously analyzing full audio signals, the DSP circuit monitors specific key parameters (power level, signal-to-noise ratio, distortion metrics) at defined intervals. The system uses threshold-based detection and simple comparison logic to determine when switching is needed, rather than continuous complex audio analysis. This substitution maintains audio quality reliability while significantly reducing processing complexity and computational requirements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4637057A1Method and system for blending audio signals
Publication Date: 2025.10.22 NXP BV
  • EP4637057A1 patent drawingFigure 1
  • EP4637057A1 patent drawingFigure 2
  • EP4637057A1 patent drawingFigure 3A

AI summary

A digital signal processing (DSP) circuit of a system for blending audio signals executes a trained machine learning model to extract audio parameters associated with audio blocks of two received audio signals and generates audio quality scores. Each audio quality score indicates an audio quality of the audio block. Upon analyzing the corresponding audio quality scores of the two audio signals, the DSP circuit outputs an audio block of one of the audio signals based on a previous blended block or blends one of the audio blocks of the two audio signals to output a blended block that includes a composition of the corresponding audio blocks of the two audio signals. The system thus outputs an audio output signal that includes such audio blocks that are associated with at least one of the two audio signals.