# US 2021/0233515 A1 · Pre-wakeword speech processing

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

- Number: US 2021/0233515 A1
- Document ID: US20210233515A1
- Status: Published application
- Kind: A1
- Application number: 17/232,609
- Filed: 2021-04-16
- Published: 2021-07-29
- Earliest claimed: 2015-03-30
- Assignees: Amazon Technologies, Inc.
- Applicants: Amazon Technologies, Inc.
- Inventors: Kurt Wesley Piersol; Gabriel Beddingfield
- Web page: https://patentgrep.com/patent/US20210233515A1

## CPC

- G10L 15/08: Speech classification or search
- G10L 17/22: Interactive procedures; Man-machine interfaces
- G10L 25/87: Detection of discrete points within a voice signal
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A system for capturing and processing portions of a spoken utterance command that may occur before a wakeword. The system buffers incoming audio and indicates locations in the audio where the utterance changes, for example when a long pause is detected. When the system detects a wakeword within a particular utterance, the system determines the most recent utterance change location prior to the wakeword and sends the audio from that location to the end of the command utterance to a server for further speech processing.

## Claims

### Claim 1

A computer-implemented method comprising:
receiving audio comprising speech;
storing, in non-transitory memory, audio data representing the audio;
determining at least a portion of the speech was directed at a device;
determining a first portion of the audio data, stored in the non-transitory memory, corresponding to the speech directed at the device; and
sending the first portion of the audio data for further processing.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising:
sending the first portion of the audio data to a first component for speech processing.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein sending the first portion of the audio data to a first component for speech processing comprises:
sending the first portion of the audio data to a remote device.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein determining at least a portion of the speech was directed at the device comprises:
determining the speech comprises a wakeword.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining a pause represented in the audio data;
determining that the pause occurs prior to the first portion of the audio data and after a second portion of the audio data; and
determining a beginning of the first portion based at least in part on the pause.

### Claim 6 (depends on 5)

The computer-implemented method of claim 5, further comprising:
refraining from sending the second portion of the audio data for speech processing.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining a pause represented in the audio data;
determining that the pause occurs after the first portion of the audio data and before a second portion of the audio data; and
determining an ending of the first portion based at least in part on the pause.

### Claim 8 (depends on 7)

The computer-implemented method of claim 7, further comprising:
refraining from sending the second portion of the audio data for speech processing.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, further comprising, after sending the first portion of the audio data for further processing:
determining further audio following the speech;
determining the further audio is not directed at the device; and
in response to determining the further audio is not directed at the device, stopping sending audio data for further processing.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, wherein determining the further audio is not directed at the device comprises:
detecting a pause following the speech.

### Claim 11

A system comprising:
at least one processor; and
at least one memory comprising instructions that, when executed by the at least one processor, cause the system to:
receive audio comprising speech;
store, in non-transitory memory, audio data representing the audio;
determine at least a portion of the speech was directed at a device;
determine a first portion of the audio data, stored in the non-transitory memory, corresponding to the speech directed at the device; and
send the first portion of the audio data for further processing.

### Claim 12 (depends on 11)

The system of claim 11, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
send the first portion of the audio data to a first component for speech processing.

### Claim 13 (depends on 12)

The system of claim 12, wherein the instructions that cause the system to send the first portion of the audio data to a first component for speech processing comprise instructions that, when executed by the at least one processor, further cause the system to:
send the first portion of the audio data to a remote device.

### Claim 14 (depends on 11)

The system of claim 11, wherein the instructions that cause the system to determine at least a portion of the speech was directed at the device comprise instructions that, when executed by the at least one processor, further cause the system to:
determine the speech comprises a wakeword.

### Claim 15 (depends on 11)

The system of claim 11, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
determine a pause represented in the audio data;
determine that the pause occurs prior to the first portion of the audio data and after a second portion of the audio data; and
determine a beginning of the first portion based at least in part on the pause.

### Claim 16 (depends on 15)

The system of claim 15, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
refrain from sending the second portion of the audio data for speech processing.

### Claim 17 (depends on 11)

The system of claim 11, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
determine a pause represented in the audio data;
determine that the pause occurs after the first portion of the audio data and before a second portion of the audio data; and
determine an ending of the first portion based at least in part on the pause.

### Claim 18 (depends on 17)

The system of claim 17, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
refrain from sending the second portion of the audio data for speech processing.

### Claim 19 (depends on 17)

The system of claim 17, wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to, after sending the first portion of the audio data for further processing:
determine further audio following the speech;
determine the further audio is not directed at the device; and
in response to determination that the further audio is not directed at the device, stop sending audio data for further processing.

### Claim 20 (depends on 19)

The system of claim 19, wherein the instructions that cause the system to determine the further audio is not directed at the device comprise instructions that, when executed by the at least one processor, further cause the system to:
detect a pause following the speech.
