Sound Source Direction Estimation Using Phase Difference Database
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech translation devices fail to accurately recognize speech in noisy environments due to noise interference, leading to increased computational complexity in estimating speaker direction.
Innovation Solution
A sound source direction estimation device that calculates phase differences from a microphone array, compares these with precalculated phase differences for various directions, and uses a feature quantity calculator to determine speech presence while inhibiting computational increase, employing a neural network for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direction estimation and noise suppression processing is performed to extract speech in noisy environments, then speech quality is improved, but the amount of computation increases
Solution Approach 1:
The system pre-calculates and stores phase difference information for multiple directions in a phase difference database before actual speech recognition occurs. During operation, the similarity calculator directly compares measured phase differences with pre-stored values, avoiding real-time complex noise suppression computations while maintaining accurate speech direction estimation
Solution Approach 2:
The patent creates a phase difference database that contains pre-computed phase difference values representing acoustic characteristics from various directions. This database serves as a reference copy that enables quick comparison and speech determination without repeating complex computational processes during actual speech recognition tasks
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The device accurately distinguishes speech in noisy environments without increasing computational load, enhancing speech recognition and translation quality by directing sound pickup in the estimated speaker direction.
Implementation Method 1
acoustic signal obtained by a microphone array which includes at least two microphone units
Implementation Method 2
calculates, from an acoustic signal obtained by a microphone array which includes at least two microphone units, a first phase difference of at least one pair of microphone units
Data Source
AI summary
A sound source direction estimation device includes: a phase difference calculator which calculates, from an acoustic signal obtained by a microphone array, a first phase difference of a pair of microphone units; a similarity calculator which calculates similarities between the calculated first phase difference and second phase differences precalculated for directions and stored in a phase difference database; a peak searcher which searches for a direction for which a highest similarity is calculated by the similarity calculator, and estimates the direction searched out to be a sound source direction; a feature quantity calculator which uses the calculated similarities, the estimated sound source direction, and an acoustic feature quantity obtained from the obtained acoustic signal, to calculate a feature quantity obtained by correcting the acoustic feature quantity; and a speech/non-speech determiner which determines whether the obtained acoustic signal indicates speech, using the feature quantity calculated by the feature quantity calculator.