Audio verification

Patent metadata

Field	Value
Publication number	US-10277581-B2
Application number	US-201514847742-A
Country	US
Kind code	B2
Filing date	Sep 8, 2015
Priority date	Sep 8, 2015
Publication date	Apr 30, 2019
Grant date	Apr 30, 2019

Abstract

One or more techniques and/or systems are provided for audio verification. An audio signal, comprising a code for user verification, may be identified. A second audio signal is created comprising speech. The audio signal and the second audio signal may be altered to comprise a same or similar volume, pitch, amplitude, and/or speech rate. The audio signal and the second audio signal may be combined to generate a verification audio signal. The verification audio signal may be presented to a user for the user verification. Verification may be performed to determine whether the user has access to content or a service based upon user input, obtained in response to the user verification audio signal, matching the code within the user verification audio signal. In an example, the user verification may comprise verifying that the user is human.

First claim

What is claimed is:

1. A system of audio verification comprising:
a processor; and
memory comprising processor-executable instructions that when executed by the processor cause implementation of an audio generation component configured to:
identify an audio signal comprising a code for user verification;
extract one or more audio segments in real-time from an on-going audio stream;
create a second audio signal comprising speech based upon the one or more audio segments extracted in real-time from the on-going audio stream;
identify a pitch and a volume of the audio signal at a first time;
identify a second pitch and a second volume of the second audio signal at a second time, wherein the pitch of the audio signal and the second pitch of the second audio signal are not within a threshold pitch similarity, and wherein the volume of the audio signal and the second volume of the second audio signal are not within a threshold volume similarity;
determine an average pitch between the pitch of the audio signal and the second pitch of the second audio signal;
determine an average volume between the volume of the audio signal and the second volume of the second audio signal;
alter the pitch of the audio signal and the second pitch of the second audio signal to be the average pitch at a third time;
alter the volume of the audio signal and the second volume of the second audio signal to be the average volume at a fourth time;
combine the audio signal and the second audio signal to generate a verification audio signal in response to:
determining that the pitch of the audio signal at the third time and the second pitch of the second audio signal at the third time are both the average pitch so that a bot is unable to discern a difference between the pitch of the audio signal and the second pitch of the second audio signal when the audio signal and the second audio signal are combined; and
determining that the volume of the audio signal at the fourth time and the second volume of the second audio signal at the fourth time are both the average volume so that a bot is unable to discern a difference between the volume of the audio signal and the second volume of the second audio signal when the audio signal and the second audio signal are combined;

present the verification audio signal to a user for the user verification, the user verification comprising verifying that the user is human; and
verify whether the user has access to content or a service based upon user input, obtained in response to the verification audio signal, matching the code within the verification audio signal.

2. The system of claim 1 , the audio generation component configured to:
identify a speaking rate of the audio signal; and
identify a second speaking rate of the second audio signal.

3. The system of claim 2 , the audio generation component configured to:
alter at least one of the speaking rate of the audio signal or the second speaking rate of the second audio signal until the speaking rate and the second speaking rate are within a threshold speaking rate similarity.

4. The system of claim 1 , the audio generation component configured to:
identify an amplitude of the audio signal; and
identify a second amplitude of the second audio signal.

5. The system of claim 4 , the audio generation component configured to:
alter at least one of the amplitude of the audio signal or the second amplitude of the second audio signal until the amplitude and the second amplitude are within a threshold amplitude similarity.

6. The system of claim 1 , the audio generation component configured to:
create the second audio signal utilizing a first audio segment and a second audio segment.

7. The system of claim 6 , the audio generation component configured to at least one of:
extract at least one of the first audio segment or the second audio segment from an audio content database; or
generate at least one of the first audio segment or the second audio segment utilizing a random speech generator.

8. The system of claim 6 , the audio generation component configured to:
randomly extract one or more portions from at least one of the first audio segment or the second audio segment; and
randomly stitch the one or more portions together to create the second audio signal.

9. The system of claim 6 , the audio generation component configured to:
randomly extract one or more portions from at least one of the first audio segment or the second audio segment;
randomly layer the one or more portions over each other to create a layered segment and a second layered segment; and
stitch the layered segment and the second layered segment together to create the second audio signal.

10. The system of claim 6 , the audio generation component configured to:
randomly extract one or more portions from at least one of the first audio segment or the second audio segment;
randomly stitch the one or more portions together to create an initial second audio signal; and
reverse the initial second audio signal to create the second audio signal.

11. The system of claim 1 , wherein the second audio signal comprises computer generated speech.

12. The system of claim 1 , the audio generation component configured to:
provide the user with an option to enter the user input audibly;
responsive to the user entering the user input audibly, identify acoustic features that are indicative of a human voice; and
responsive to the acoustic features indicating that the user input was spoken by the human voice, verify the user access to the content or the service.

13. The system of claim 1 , the audio generation component configured to:
provide the user an option to enter the user input audibly;
responsive to the user entering the user input audibly, identify acoustic features that are indicative of a human voice; and
responsive to the acoustic features indicating the user input was not spoken by a human voice, deny the user access to the content or the service.

14. A method of audio verification comprising:
identifying an audio signal comprising a code for user verification;
extracting one or more audio segments in real-time from an on-going audio stream;
creating a second audio signal comprising speech based upon the one or more audio segments extracted in real-time from the on-going audio stream;
identifying a speaking rate and an amplitude of the audio signal at a first time;
identifying a second speaking rate and a second amplitude of the second audio signal at a second time, wherein the speaking rate of the audio signal and the second speaking rate of the second audio signal are not within a threshold speaking rate similarity, and wherein the amplitude of the audio signal and the second amplitude of the second audio signal are not within a threshold amplitude similarity;
altering the speaking rate of the audio signal and the second speaking rate of the second audio signal by altering the speaking rate be more similar to the second speaking rate at the second time and altering the second speaking rate to be more similar to the speaking rate at the first time until the speaking rate and the second speaking rate are within the threshold speaking rate similarity at a third time;
altering the amplitude of the audio signal and the second amplitude of the second audio signal by altering the amplitude be more similar to the second amplitude at the second time and altering the second amplitude to be more similar to the amplitude at the first time until the amplitude and the second amplitude are within the threshold amplitude similarity at a fourth time;
combining the audio signa

Assignees

• Oath Inc

Inventors

• Chandrasekharan Manjana
• Horiguchi Keiko
• Stent Amanda Joy
• Baeza-Yates Ricardo Alberto
• Kuwano Jeffrey
• Thomas Achint Oommen
• Chang Yi

Classifications

• G10L17/06

  Decision making techniques; Pattern matching strategies · CPC title

• G06F3/165

  Management of the audio stream, e.g. setting of volume, audio stream path · CPC title

• G06F2221/2133

  Verifying human interaction, e.g., Captcha · CPC title

• G06F3/167

  Audio in a user interface, e.g. using voice commands for navigating, audio feedback · CPC title

• H04L9/3226

  using a predetermined code, e.g. password, passphrase or PIN (network architectures or network communication protocols for supporting authentication of entities using passwords in a packet data network H04L63/083) · CPC title