Echo Suppression Device Double-Talk Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech signal processing devices face difficulties in accurately detecting the double-talk state and suppressing echoes when the signal level of the transmitting side is low, leading to ineffective echo suppression.
Innovation Solution
An echo suppression device that uses a frequency mask generated from a learning signal to compare with the input signal spectrum for each frequency band, enabling accurate double-talk detection and echo suppression by differentiating between near-end voice and residual echo, even when the residual echo has higher power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If signal level of transmitting side is low, then near-end voice can be captured, but double-talk detection accuracy deteriorates
Solution Approach 1:
The patent transitions from time-domain signal level comparison to frequency-domain spectrum analysis. By comparing power spectra or amplitude spectra across multiple frequency bands, the system can detect double-talk states even when overall signal levels are low, as the frequency distribution characteristics provide additional discrimination dimensions beyond simple power comparison
Solution Approach 2:
The patent changes the detection parameter from time-domain signal level to frequency-domain spectrum characteristics. By analyzing the distribution of power across frequency bands and comparing spectral shapes, the system achieves more robust double-talk detection that is not degraded by low transmitting signal levels
2Object-generated harmful factors
If echo suppressor is activated to suppress residual echo, then echo is reduced, but near-end voice may be degraded
Solution Approach 1:
The patent applies different processing strategies to different frequency bands based on local spectral characteristics. By identifying which frequency bands contain near-end voice and which contain primarily echo, the echo suppressor can selectively attenuate echo in specific frequency regions while preserving near-end voice in other regions, achieving local optimization of voice quality and echo suppression
Data Source
AI summary
A double-talk state can be accurately detected, and based on a detection result, echo can be appropriately suppressed. When a sound is output from a speaker and only the output sound is input to a microphone, a comparison is made, for each of different frequency bands, between a frequency mask generated based on a power spectrum or an amplitude spectrum for a learning signal transmitted through a transmitting signal path and a value of a power spectrum or an amplitude spectrum for an input signal input from the microphone, to detect whether there is a double-talk state. In a case of detecting that no signal is being transmitted through the transmitting signal path and that a signal is being transmitted through the receiving signal path, an echo suppressor is used to execute processing of suppressing an echo in the input signal.