# US 12,112,764 B2 · Delay estimation using frequency spectral descriptors

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,112,764 B2
- Document ID: US12112764B2
- Status: Granted
- Kind: B2
- Application number: 17/823,521
- Filed: 2022-08-31
- Published: 2024-02-29
- Granted: 2024-10-08
- Earliest claimed: 2022-08-31
- Assignees: Nuvoton Technology Corporation
- Applicants: Nuvoton Technology Corporation
- Inventors: Powen Ru; Dung Nguyen; Andrew Zamansky
- Web page: https://patentgrep.com/patent/US12112764B2

## CPC

- G10L 19/02: using spectral analysis, e.g. transform vocoders or subband vocoders
- G10L 25/48: specially adapted for particular use
- G10L 2021/02082: {the noise being echo, reverberation of the speech}
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method is disclosed to estimate the delay between an original signal and the corresponding captured signal. The signals are transformed and buffered to two sets of spectral descriptors for a similarity measure. The method advantageously offers robust delay estimation for inconsistent delays and adverse spectral distortions.

## Claims

### Claim 1

A system comprising:
a host device to provide a known waveform;
a signal transmitter to obtain the known waveform from the host device via a channel and to emit a signal corresponding to the known waveform; and
a signal receiver to convert the signal to a received waveform and emit the received waveform to the host device;
wherein the host device comprises a processor being configured to:
transform the known waveform to a reference spectral descriptor matrix and a reference magnitude representation matrix;
transform the received waveform via the signal receiver to a received spectral descriptor matrix;
obtain a similarity measure between the reference spectral descriptor matrix and the received spectral descriptor matrix;
accumulate the similarity measure based on at least one statistic of the reference magnitude representation matrix to obtain a cumulative similarity measure;
determine a delay based on the cumulated similarity measure; and
output information characterizing the delay.

### Claim 2 (depends on 1)

The system of claim 1, wherein the known waveform is an audio content, the signal transmitter is a loudspeaker, the signal is an acoustic signal, and the signal receiver is a microphone.

### Claim 3 (depends on 1)

The system of claim 1, wherein the channel is a wired channel including one of High-Definition Multimedia Interface (HDMI) and Universal Serial Bus (USB).

### Claim 4 (depends on 1)

The system of claim 1, wherein the channel is a wireless channel including one of Bluetooth and WiFi.

### Claim 5 (depends on 1)

The system of claim 1, wherein the processor is configured to convert the known waveform to a first spectrum, add a floor to the first spectrum, convert the floor-added first spectrum to a first logarithmic spectrum, convert the first logarithmic spectrum to a first series of coefficients via a transformation method, wherein less than 30% of the first series of coefficients are used as reference spectral descriptors to represent the known waveform; and
wherein the processor is configured to convert the received waveform to a second spectrum, add the floor to the second spectrum, convert the floor-added second spectrum to a second logarithmic spectrum, convert the second logarithmic spectrum to a second series of coefficients via the transformation method, wherein less than 30% of the second series of coefficients are used as received spectral descriptors to represent the received waveform.

### Claim 6 (depends on 5)

The system of claim 5, wherein the transformation method is discrete cosine transform (DCT).

### Claim 7 (depends on 5)

The system of claim 5, wherein the transformation method is one of discrete sine transform (DST), cepstrum, principal component analysis (PCA), and wavelet transform (WT).

### Claim 8 (depends on 1)

The system of claim 1, wherein the reference magnitude representation matrix is a root-mean-square (RMS) of the known waveform.

### Claim 9 (depends on 1)

The system of claim 1, wherein the reference magnitude representation matrix is a maximum magnitude, an average magnitude, a power, or a sound pressure level (SPL) of the known waveform.

### Claim 10 (depends on 1)

The system of claim 1, wherein the similarity measure is cross-correlation.

### Claim 11 (depends on 1)

The system of claim 1, wherein the similarity measure is distance.

### Claim 12 (depends on 1)

The system of claim 1, wherein the at least one statistic is minimum, average, or sum.

### Claim 13 (depends on 10)

The system of claim 10, wherein a candidate delay with maximum cumulated cross-correlation is determined as the delay.

### Claim 14 (depends on 11)

The system of claim 11, wherein a candidate delay with minimum cumulated distance is determined as the delay.

### Claim 15

A computer-implemented method comprising:
transforming a known waveform to a reference spectral descriptor matrix and storing the reference spectral descriptor matrix in a first buffer;
transforming a received waveform to a received spectral descriptor matrix and storing the received spectral descriptor matrix in a second buffer;
transforming the known waveform to a reference magnitude representation matrix and storing the reference magnitude representation matrix in a third buffer;
obtaining a similarity measure between the reference spectral descriptor matrix and the received spectral descriptor matrix;
accumulating the similarity measure based on at least one statistic of the reference magnitude representation matrix to obtain a cumulative similarity measure;
determining a delay based on the cumulated similarity measure; and
outputting information characterizing the delay.

### Claim 16 (depends on 15)

The method of claim 15, wherein the method is configured to convert the known waveform to a first spectrum, add a floor to the first spectrum, convert the floor-added first spectrum to a first logarithmic spectrum, convert the first logarithmic spectrum to a first series of coefficients via a transformation method, wherein less than 30% of the first series of coefficients are used as reference spectral descriptors to represent the known waveform; and
wherein the method is configured to convert the received waveform to a second spectrum, add the floor to the second spectrum, convert the floor-added second spectrum to a second logarithmic spectrum, convert the second logarithmic spectrum to a second series of coefficients via the transformation method, wherein less than 30% of the second series of coefficients are used as received spectral descriptors to represent the received waveform.

### Claim 17 (depends on 16)

The method of claim 16, wherein the transformation method is discrete cosine transform (DCT).

### Claim 18 (depends on 15)

The method of claim 15, wherein the reference magnitude representation matrix is a root-mean-square (RMS) of the known waveform.

### Claim 19 (depends on 15)

The method of claim 15, wherein the similarity measure is cross-correlation, and a candidate delay with maximum cumulated cross-correlation is determined as the delay.

### Claim 20 (depends on 15)

The method of claim 15, wherein the similarity measure is distance, and a candidate delay with minimum distance is determined as the delay.