Audio data receipt/exposure measurement with code monitoring and signature extraction

Abstract

Systems and methods are provided for gathering audience measurement data relating to receipt of and / or exposure to audio data by an audience member. Audio data is monitored to detect a monitoring code. Based on detection of the monitoring code, a signature characterizing the audio data is extracted.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S) [0001] This application is a continuation of nonprovisional patent application Ser. No. 10 / 256,834, filed Sep. 27, 2002 and assigned to the assignee of the present invention, and which is hereby incorporated by reference in its entirety.FIELD OF THE INVENTION [0002] The invention relates to systems and methods for gathering data reflecting receipt of, and / or exposure to, audio data by detecting a code in, and extracting a signature from, audio data. BACKGROUND OF THE INVENTION [0003] There is considerable interest in identifying and / or measuring the receipt of, and or exposure to, audio data by an audience in order to provide market information to advertisers, media distributors, and the like, to verify airing, to calculate royalties, to detect piracy, and for any other purposes for which an estimation of audience receipt or exposure is desired. [0004] The emergence of multiple, overlapping media distribution pathways, as well as the wide va...

AI Technical Summary

Problems solved by technology

The emergence of multiple, overlapping media distribution pathways, as well as the wide variety of available user systems (e.g. PC's, PDA's, portable CD players, Internet, appliances, TV, radio, etc.) for receiving audio data, has greatly complicated the task of measuring audience receipt of, and exposure to, individual program segments.

However, one disadvantage of using such pattern matching techniques is that, because there is no predetermined point in the program signal from which signature extraction is designated to begin, each program signal must continually undergo signature extraction, and each of these many successive signatures extracted from a single program signal must be compared to each of the reference signatures in the database.

This, of course, requires a tremendous amount of data processing, which, due to the ever increasing methods and amounts of audio data transmission, is becoming more and more economically impractical.

One disadvantage of this technique, however, is that the presence of a code that triggers the extraction of a signature from a portion of the signal before or after the portion of the signal that has been encoded necessarily limits the amount of information that can be obtained for producing the signature, as the encoded portion itself may contain information useful for producing the signature, and moreover, may contain information required to measure the values of certain features, such as changes of certain properties or ratios over time, which might not be accurately measured when a temporal segment of the signal (i.e. the encoded portion) cannot be used.

Another disadvantage of this technique is that, because the trigger code is of short duration, the likelihood of its detection is reduced.

One disadvantage of such short codes is the diminished probability of detection that may result when a signal is distorted or obscured, as is the case when program signals are broadcast in acoustic environments.

In such environments, which often contain significant amounts of noise, the trigger code will often be overwhelmed by noise, and thus, not be detected.

Yet another specific disadvantage of such short codes is the diminished probability of detection that may result when certain portions of a signal are unrecoverable, such as when burst errors occur during transmission or reproduction of encoded audio signals.

Burst errors may appear as temporally contiguous segments of signal error.

Such errors generally are unpredictable and substantially affect the content of an encoded audio signal.

Burst errors typically arise from failure in a transmission channel or reproduction device due to external interferences, such as overlapping of signals from different transmission channels, an occurrence of system power spikes, an interruption in normal operations, an introduction of noise contamination (intentionally or otherwise), and the like.

In a transmission system, such circumstances may cause a portion of the transmitted encoded audio signals to be entirely unreceivable or significantly altered.

Absent retransmission of the encoded audio signal, the affected portion of the encoded audio may be wholly unrecoverable, while in other instances, alterations to the encoded audio signal may render the embedded information signal undetectable.

In systems for acoustically reproducing audio signals recorded on media, a variety of factors may cause burst errors in the reproduced acoustic signal.

Commonly, an irregularity in the recording media, caused by damage, obstruction, or wear, results in certain portions of recorded audio signals being irreproducible or significantly altered upon reproduction.

Also, misalignment of, or interference with, the recording or reproducing mechanism relative to the recording medium can cause burst-type errors during an acoustic reproduction of recorded audio signals.

Further, the acoustic limitations of a speaker as well as the acoustic characteristics of the listening environment may result in spatial irregularities in the distribution of acoustic energy.

Such irregularities may cause burst errors to occur in received acoustic signals, interfering with recovery of the trigger code.

A further disadvantage of this technique is that reproduction of a single, short-lived code that triggers signature extraction does not reflect the receipt of a signal by any audience member who was exposed to part, or even most, of the signal if the audience member was not present at the precise point at which the portion of the signal containing the trigger code was broadcast.

Yet another disadvantage of this technique is that a single code of short duration that triggers signature extraction does not provide any data reflecting the amount of time for which an audience member was exposed to the audio data.

Still another disadvantage of this technique is that a single code that triggers signature extraction cannot mark “beginning” and “end” portions of a program segment, which may be desired, for example, to determine the time boundaries of the segment.

Images