# US 2021/0343294 A1 · System and method for speaker identification in audio data

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

- Number: US 2021/0343294 A1
- Document ID: US20210343294A1
- Status: Published application
- Kind: A1
- Application number: 16/864,856
- Filed: 2020-05-01
- Published: 2021-11-04
- Earliest claimed: 2020-05-01
- Assignees: Raytheon Company
- Applicants: Raytheon Company
- Inventors: Jonathan C. Wintrode; Nicholas J. Hinnerschitz; Aleksandr R. Jouravlev
- Web page: https://patentgrep.com/patent/US20210343294A1

## CPC

- G10L 17/02: Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
- G10L 25/51: for comparison or discrimination
- G06F 16/683: using metadata automatically derived from the content
- G06N 3/045: Combinations of networks
- G06N 3/0499: Feedforward networks
- G06N 3/09: Supervised learning
- G10L 17/00: Speaker identification or verification techniques
- G10L 17/06: Decision making techniques; Pattern matching strategies
- G10L 17/18: Artificial neural networks; Connectionist approaches
- G10H 2240/141: Library retrieval matching, i.e. any of the steps of matching an inputted segment or phrase with musical database contents, e.g. query by humming, singing or playing; the steps may include, e.g. musical analysis of the input, musical feature extraction, query formulation, or details of the retrieval process
- G10H 2250/311: Neural networks for electrophonic musical instruments or musical processing, e.g. for musical recognition or control, automatic composition or improvisation
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G10H: ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE

## Abstract

A system for identifying audio data includes a feature extraction module receiving unknown input audio data and dividing the unknown input audio data into a plurality of segments of unknown input audio data. A similarity module receives the plurality of segments of the unknown input audio data and receives known audio data from a known source, the known audio data being divided into a plurality of segments of known audio data. The similarity module performs comparisons between the segments of unknown input audio data and respective segments of known audio data and generates a respective plurality of similarity values representative of similarity between the segments of the comparisons, the comparisons being performed serially. The similarity module terminates the comparisons if the similarity values indicate insufficient similarity between the segments of the comparisons, prior to completing comparisons for all segments of the unknown input audio data.

## Claims

### Claim 1

A system for identifying audio data, comprising:
a feature extraction module for receiving an unknown input audio data stream from a source to be identified and dividing the unknown input audio data stream into a plurality of segments of unknown input audio data; and
a similarity module for receiving the plurality of segments of the unknown input audio data and for receiving a known audio data stream from a known source, the known audio data stream being divided into a plurality of segments of known audio data, the similarity module performing comparisons between the segments of unknown input audio data and respective segments of known audio data and generating a respective plurality of similarity values representative of similarity between the segments of the comparisons, the comparisons being performed serially, the similarity module terminating the comparisons if the similarity values indicate insufficient similarity between the segments of the comparisons, prior to completing comparisons for all segments of the unknown input audio data.

### Claim 2 (depends on 1)

The system of claim 1, wherein each segment of unknown audio data comprises an unknown data vector comprising a plurality of data values associated with the segment of unknown input audio data; and each segment of known audio data comprises a known data vector comprising a plurality of data values associate with the segment of known audio data.

### Claim 3 (depends on 2)

The system of claim 2, wherein the similarity module, in generating each similarity value, computes a vector distance between the unknown data vector and the known data vector of the segments of the comparison being performed.

### Claim 4 (depends on 1)

The system of claim 1, wherein the feature extraction module comprises a deep neural network.

### Claim 5 (depends on 1)

The system of claim 1, wherein the similarity module performs a probabilistic linear discriminant analysis (PLDA) in generating the plurality of similarity values.

### Claim 6 (depends on 1)

The system of claim 1, wherein extents of each segment are defined by a time duration.

### Claim 7 (depends on 1)

The system of claim 1, wherein extents of each segment are defined by a quantity of data in the segment.

### Claim 8

A method for identifying audio data, comprising:
receiving, at a feature extraction module, an unknown input audio data stream from a source to be identified and dividing the unknown input audio data stream into a plurality of segments of unknown input audio data; and
receiving, at a similarity module, the plurality of segments of the unknown input audio data and receiving, at the similarity module, a known audio data stream from a known source, the known audio data stream being divided into a plurality of segments of known audio data, the similarity module performing comparisons between the segments of unknown input audio data and respective segments of known audio data and generating a respective plurality of similarity values representative of similarity between the segments of the comparisons, the comparisons being performed serially, the similarity module terminating the comparisons if the similarity values indicate insufficient similarity between the segments of the comparisons, prior to completing comparisons for all segments of the unknown input audio data.

### Claim 9 (depends on 8)

The method of claim 8, wherein each segment of unknown audio data comprises an unknown data vector comprising a plurality of data values associated with the segment of unknown input audio data; and each segment of known audio data comprises a known data vector comprising a plurality of data values associate with the segment of known audio data.

### Claim 10 (depends on 9)

The method of claim 9, wherein the similarity module, in generating each similarity value, computes a vector distance between the unknown data vector and the known data vector of the segments of the comparison being performed.

### Claim 11 (depends on 8)

The method of claim 8, wherein the feature extraction module comprises a deep neural network.

### Claim 12 (depends on 8)

The method of claim 8, wherein the similarity module performs a probabilistic linear discriminant analysis (PLDA) in generating the plurality of similarity values.

### Claim 13 (depends on 8)

The method of claim 8, wherein extents of each segment are defined by a time duration.

### Claim 14 (depends on 8)

The method of claim 8, wherein extents of each segment are defined by a quantity of data in the segment.
