Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
2000-02-18
2003-02-18
McFadden, Susan (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S270000
Reexamination Certificate
active
06523008
ABSTRACT:
FIELD OF THE INVENTION
The invention relates generally to lie detection, and more particularly relates to truth-enabling Internet communications.
BACKGROUND OF THE INVENTION
Text messages are ubiquitous on the Internet. For example, users of the Internet send billions of e-mail messages every day. Over the Web, people can also send text messages that are not included in e-mails. Text messages that people may transmit over the Web include text messages posted on message boards, text messages transmitted between users of online dating services, and online auction ads.
There has heretofore been no convenient and reliable way for a recipient to determine if a sender's text message is truthful. Because the sender of a text message is not orally presenting the information included in the text message in person, the recipient has no opportunity to observe the sender's demeanor while making certain representations or to easily question the sender about the representations the sender has made. This problem is particularly acute where a recipient must rely on representations in a text message made by a sender that the recipient does not personally know—a situation that arises frequently on the Internet.
For obvious reasons, enabling a recipient of a text message to determine the veracity of that text message is desirable. Therefore, there is a need in the art for enabling a text message recipient to determine the truthfulness of a received text message. More generally, there is a need in the art for truth-enabling Internet communications.
SUMMARY OF THE INVENTION
The present invention meets the needs described above in methods for truth-enabling communications over a computer network using computer voice stress analysis. In a typical embodiment, a sender sends a text message over a computer network to a recipient and creates speech input indicating the truthfulness of the text message. A computer voice stress analysis program analyzes the speech input and advantageously informs the recipient of the truthfulness of the text message.
By applying computer voice stress analysis to speech input in order to determine the truthfulness of the text message, the present invention provides the advantage of requiring minimal additional effort on the behalf of either the sender or the recipient to truth-enable the text message. Furthermore, the present invention also provides the advantage of being operable using standard computer equipment which users of a computer linked to a computer network typically have. Though the sender must have a microphone or other audio input peripheral in order to create the speech input indicating the truthfulness of the text message, microphones are becoming increasingly common in personal computer configurations.
Generally, computer voice stress analysis is a process for determining the truthfulness of speech input that includes assessing aspects of the speech input which may be undetectable to the human ear. When a speaker lies, for example, the speaker typically experiences emotional and physical stress. This stress often manifests itself in aspects of the speech input, such as microtremors in portions of the speech input corresponding to the lie. Though people typically cannot hear these microtremors, a computer voice stress analysis program may nonetheless discover them. Searching for such microtremors helps the computer voice stress analysis program to assess the likelihood that the speaker is lying. The computer voice stress analysis program may also examine other indicators of stress and deception in the speech input to help assess the likelihood that the speaker is lying.
In one embodiment of the present invention, a sender creates digital data, such as a text message, at a first computer of the computer network. At some point, the first computer transmits the text message over the computer network to a recipient, who may be an individual at a second computer of the computer network. The first computer may transmit the text message directly to the recipient.
Alternatively, the first computer may send the text message via an intermediary computer server which offers additional functionality corresponding to the nature of the text message. Such an intermediary computer server may comprise an e-mail application service provider (the text message is included in an e-mail addressed to the recipient), a message board (the text message comprises a message to be posted on the message board and which may be read by a recipient that the sender does not personally know), an online auction site (the text message comprises an ad for an item the sender wants to sell through the online auction site, and the recipient is a potential bidder on the item), a site which introduces singles to each other and allows them to exchange text messages, or any other computer server site through which a sender may send a text message. To transmit the text message via an intermediary computer server, the sender's computer first sends the text message to the intermediary computer server, and the intermediary computer server subsequently sends the text message to the recipient's computer for review by the recipient.
In addition to creating the text message, the sender also enters into the first computer speech input which indicates the veracity of the text message. For example, the speech input may indicate the veracity of the text message through an explicit assertion that the sender believes the text message to be truthful. Speech input comprising a vocalization by the sender of a portion of the text message also indicates the veracity of the text message because computer voice stress analysis can later determine if the vocalized portion of the text message is truthful. Speech input comprising such a vocalization of the text message therefore implicitly asserts that the vocalized portion of the text message is truthful.
By applying computer voice stress analysis to the speech input, a computer voice stress analysis program determines the veracity of the text message. The computer voice stress analysis program may be located at the sender's computer, at the recipient's computer, or at an intermediary computer server. The computer voice stress analysis program may next provide the speech input and an indication of the veracity of the speech input to the recipient. The recipient can then listen to the speech input to determine the applicability of the speech input to the text message. The recipient can thereby assess the truthfulness of the text message.
Once the computer voice stress analysis program has determined the veracity of the text message, a computer having this information need not provide an indication of the veracity of the text message to the recipient. Instead, a computer knowing the truthfulness of the text message may forward the text message to the recipient only if the text message is truthful.
Various embodiments of the present invention use speech recognition to verify the content of the speech input. Specifically, well known speech recognition techniques can determine the words intended by the speech input or verify that the words corresponding to the speech input match words the computer expects the sender to vocalize. The speech recognition program may then inform the recipient about whether or not the content of the speech input matches an expected content so that the recipient need not listen to the speech input to determine the applicability of the speech input to the text message.
Generally described, the present invention comprises a method for truth-enabling communications over a computer network. A computer network receives digital data from a sender at a first computer of the computer network. The computer network also receives from the sender an oral certification statement including at least one assertion of fact concerning the digital data. A computer voice stress analysis program determines the veracity of the certification statement using computer voice stress analysis. The computer network then transmits the digital data to
Avrunin Adam
DelGiorno Matthew
Avrunin Adam
McFadden Susan
LandOfFree
Method and system for truth-enabling internet communications... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for truth-enabling internet communications..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for truth-enabling internet communications... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3172189