ADJUSTING SPEED OF HUMAN SPEECH PLAYBACK

• US 20190318758A1
• Filed: 04/25/2019
• Published: 10/17/2019
• Est. Priority Date: 08/15/2017
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Ma, Zhaoqing, Hardie, Tony Roy, Devaraj, Christo Frank
• Granted Patent: US 11,232,808 B2
• Time in Patent Office: Days
• CPC Class Codes: G10L 21/04 Time compression or expansion
• CPC Class Codes: G10L 25/27 characterised by the analysis technique
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10

Abstract

A system configured to vary a speech speed of speech represented in input audio data without changing a pitch of the speech. The system may vary the speech speed based on a number of different inputs, including non-audio data, data associated with a command, or data associated with the voice message itself. The non-audio data may correspond to information about an account, device or user, such as user preferences, calendar entries, location information, etc. The system may analyze audio data associated with the command to determine command speech speed, identity of person listening, etc. The system may analyze the input audio data to determine a message speech speed, background noise level, identity of the person speaking, etc. Using all of these inputs, the system may dynamically determine a target speech speed and may generate output audio data having the target speech speed.

Claims

1-20. -20. (canceled)

21. A computer-implemented method, comprising:

receiving input audio data representing input speech;

determining an input speech speed corresponding to the input audio data;

determining output data based at least on the input audio data;

determining a target output speed based at least on the input speech speed;

using the output data to generate output audio data representing output speech, the output speech corresponding to the target output speed; and

causing a device to output the output audio data.

22. The computer-implemented method of claim 21, further comprising:

determining preference data corresponding to the input speech, the preference data representing at least one of a previously selected target output speed, a previously used target output speed, or location data associated with the preference data, and

wherein the target output speed is determined further based at least in part on the preference data.

23. The computer-implemented method of claim 21, further comprising:

determining urgency data associated with a user profile corresponding to the input speech, the urgency data representing at least one of location data associated with the user profile, calendar data associated with the user profile, or incoming communication data associated with the user profile, and

wherein the target output speed is determined further based at least in part on the urgency data.

24. The computer-implemented method of claim 21, further comprising:

determining the target output speed corresponding to a first portion of the output data;

determining a second target output speed corresponding to a second portion of the output data; and

wherein a first portion of the output speech corresponds to the target output speed and a second portion of the output speech corresponds to the second target output speed.

25. The computer-implemented method of claim 24, further comprising:

determining a difference between the target output speed and the second target output speed;

dividing the difference by a maximum transition value to determine a number of increments; and

determining one or more intermediate target output speeds corresponding to a third portion of the first output audio data, the third portion being between the first portion and the second portion, a number of the one or more intermediate target output speed corresponding to the number of increments,

wherein the first portion of the output speech corresponds to the target output speed, the second portion of the output speech corresponds to the second target output speed, and a third portion of the output speech corresponds to the one or more intermediate target output speeds.

26. The computer-implemented method of claim 21, further comprising:

determining an input volume level associated with the input speech;

determining a target volume level based at least in part on the input volume level; and

associating the target volume level with the output audio data.

27. The computer-implemented method of claim 21, wherein:

receiving the input audio data comprises receiving a voice command to play a voice message;

the output data represents audio data corresponding to the voice message;

the method further comprises determining a message speech speed associated with the output data; and

determining the target output speed comprises determining the target output speed based at least in part on the input speech speed and the message speech speed.

28. The computer-implemented method of claim 27, further comprising:

determining a user profile corresponding to the voice command;

determining playback speed preferences associated with the user profile;

determining configuration data corresponding to information about at least one of the user profile, the voice command or a number of voice messages associated with the user profile; and

determining quality data corresponding to an audio quality of the output data, and

wherein determining the target output speed comprises determining the target output speed based at least in part on one of the input speech speed, the message speech speed, the configuration data or the quality data.

29. The computer-implemented method of claim 27, further comprising:

determining a first user profile corresponding to the voice command;

determining first preference data associated with the first user profile, the first preference data indicating at least one of a previously selected target output speed, a previously used target output speed, or location data associated with the first user profile;

determining a second user profile corresponding to the voice message; and

determining second preference data associated with the second user profile, the second preference data indicating at least one of a preferred output speed for the voice message,

wherein determining the target output speed comprises determining the target output speed based at least in part on one of the input speech speed, the message speech speed, the first preference data or the second preference data.

30. The computer-implemented method of claim 21, further comprising:

determining a plurality of positions in the output data in which to insert a duration of silence, the plurality of positions including a first position; and

generating the output audio data using the output data, the output audio data including the duration of silence at the first position.

31. The computer-implemented method of claim 21, wherein:

the output data includes a representation of first speech associated with a first user profile and a representation of second speech associated with a second user profile, wherein the message speech speed is associated with the first speech and the target output speed is associated with the first speech, and

the method further comprises;

determining a second message speech speed associated with the second speech;

determining a second target output speed corresponding to the second speech; and

using the output data to generate the output audio data representing the output speech, a first portion of the output speech corresponding to the target output speed and a second portion of the output speech corresponding to the second target output speed.

32. A system comprising:

at least one processor; and

memory including instructions operable to be executed by the at least one processor to configure the system to;

receive input audio data representing input speech;

determine an input speech speed corresponding to the input audio data;

determine output data based at least on the input audio data;

determine a target output speed based at least on the input speech speed;

using the output data to generate output audio data representing output speech, the output speech corresponding to the target output speed; and

cause a device to output the output audio data.

33. The system of claim 32, wherein the memory further includes instructions that, when executed, further configure the system to:

determine preference data corresponding to the input speech, the preference data representing at least one of a previously selected target output speed, a previously used target output speed, or location data associated with the preference data, and

wherein the target output speed is determined further based at least in part on the preference data.

34. The system of claim 32, wherein the memory further includes instructions that, when executed, further configure the system to:

determine urgency data associated with a user profile corresponding to the input speech, the urgency data representing at least one of location data associated with the user profile, calendar data associated with the user profile, or incoming communication data associated with the user profile, and

wherein the target output speed is determined further based on at least in part the urgency data.

35. The system of claim 32, wherein the memory further includes instructions that, when executed, further configure the system to:

determine the target output speed corresponding to a first portion of the output data;

determine a second target output speed corresponding to a second portion of the output data;

wherein a first portion of the output speech corresponds to the target output speed and a second portion of the output speech corresponds to the second target output speed.

36. The system of claim 35, wherein the memory further includes instructions that, when executed, further configure the system to:

determine a difference between the target output speed and the second target output speed;

divide the difference by a maximum transition value to determine a number of increments;

determine one or more intermediate target output speeds corresponding to a third portion of the first output audio data, the third portion being between the first portion and the second portion, a number of the one or more intermediate target output speeds corresponding to the number of increments; and

wherein the first portion of the output speech corresponds to the target output speed, the second portion of the output speech corresponds to the second target output speed, and a third portion of the output speech corresponds to the one or more intermediate target output speeds.

37. The system of claim 32, wherein the memory further includes instructions that, when executed, further configure the system to:

determine an input volume level associated with the input speech;

determine a target volume level based at least in part on the input volume level;

associate the target volume level with the output audio data.

38. The system of claim 32, wherein:

the instruction to receive the input audio data further configures the system to receive a voice command to play a voice message,

the output data represents audio data corresponding to the voice message,

the memory further includes instructions that, when executed, further configure the system to determine a message speech speed associated with the third output audio data, and

the instruction to determine the target output speed further configures the system to determine the target output speed based at least in part on the input speech speed and the message speech speed.

39. The system of claim 38, wherein the memory further includes instructions that, when executed, further configure the system to:

determine a user profile associated with the voice command;

determine playback speed preferences associated with the user profile;

determine configuration data corresponding to information about at least one of the user profile, the voice command or a number of voice messages associated with the user profile; and

determine quality data corresponding to an audio quality of the third output audio data, and

wherein the instructions that configure the system to determine the target output speed further configure the system to determine the target output speed based at least in part on one of the input speech speed, the message speech speed, the configuration data or the quality data.

40. The system of claim 38, wherein the memory further includes instructions that, when executed, further configure the system to:

determine a first user profile corresponding to the voice command;

determine first preference data associated with the first user profile, the first preference data indicating at least one of a previously selected target output speed, a previously used target output speed, or location data associated with the first user profile;

determine a second user profile corresponding to the voice message; and

determine second preference data associated with the second user profile, the second preference data indicating at least one of a preferred output speed for the voice message, and

wherein the instructions that configure the system to determine the second target output speed further configure the system to determine the second target output speed based at least in part on one of the second input speech speed, the message speech speed, the first preference data or the second preference data.