ABC/HR: Difference between revisions
(Many new things) |
m (→Difference with [[ABX]]: changed "subjective bias" to "personal bias" as with previous minor edit in previous section) |
||
| (3 intermediate revisions by 2 users not shown) | |||
| Line 3: | Line 3: | ||
It is a method to 'score' the quality of audio encodings by comparing them alongside a reference sample. The 'hidden' comes from the fact that the listener does not know which sample -- the left or the right one -- is the reference. | It is a method to 'score' the quality of audio encodings by comparing them alongside a reference sample. The 'hidden' comes from the fact that the listener does not know which sample -- the left or the right one -- is the reference. | ||
The reference used usually is a sample of higher quality than the tested samples, e.g. uncompressed/[[losslessly | The reference used usually is a sample of higher quality than the tested samples, e.g. uncompressed/[[lossless|losslessly]] compressed audio tracks, or a [[lossy|lossily]] compressed track with higher [[bitrate]]. | ||
==Purpose== | ==Purpose== | ||
Blind comparison and blind quality rating to remove the effects of | Blind comparison and blind quality rating to remove the effects of personal bias and the [[placebo effect]]. | ||
==Difference with [[ABX]]== | ==Difference with [[ABX]]== | ||
[[ABX]] is used to detect audible differences blindly, thereby removing | [[ABX]] is used to detect audible differences blindly, thereby removing personal bias or the placebo effect and, over multiple trials, estimating the probability that the tester was guessing. | ||
While an ABC/HR tool can do the same, it adds to that the ability to provide a quality rating on a standardized scale, and multiple participants' results can be statistically evaluated to estimate error bars and statistical significance of differences between encoders or encoder parameters in ranking their quality or tying them. | While an ABC/HR tool can do the same, it adds to that the ability to provide a quality rating on a standardized scale, and multiple participants' results can be statistically evaluated to estimate error bars and statistical significance of differences between encoders or encoder parameters in ranking their quality or tying them. | ||
| Line 17: | Line 17: | ||
===Standardized quality scale=== | ===Standardized quality or impairment scale=== | ||
It is most common to use the 1.0 to 5.0 scale defined by [http://www.hydrogenaudio.org/forums/index.php?showtopic=53107 ITU-R BS.1116]. Any value (including fractions) between 1.0 and 5.0 is valid, with the exact whole number representing the following definitions to describe that degree of impairment: | |||
* 5.0 : Imperceptible | |||
* 4.0 : Perceptible, but not annoying | |||
* 3.0 : Slightly annoying | |||
* 2.0 : Annoying | |||
* 1.0 : Very annoying | |||
==References== | ==References== | ||
Latest revision as of 21:20, 15 May 2007
ABC/HR is an abbreviation of ABC/Hidden Reference.
It is a method to 'score' the quality of audio encodings by comparing them alongside a reference sample. The 'hidden' comes from the fact that the listener does not know which sample -- the left or the right one -- is the reference.
The reference used usually is a sample of higher quality than the tested samples, e.g. uncompressed/losslessly compressed audio tracks, or a lossily compressed track with higher bitrate.
Purpose
Blind comparison and blind quality rating to remove the effects of personal bias and the placebo effect.
Difference with ABX
ABX is used to detect audible differences blindly, thereby removing personal bias or the placebo effect and, over multiple trials, estimating the probability that the tester was guessing.
While an ABC/HR tool can do the same, it adds to that the ability to provide a quality rating on a standardized scale, and multiple participants' results can be statistically evaluated to estimate error bars and statistical significance of differences between encoders or encoder parameters in ranking their quality or tying them.
ABC/HR tends to find particular application in low-to-medium bitrate listening tests (below the quality expected to constitute transparency). As part of a well-designed listening test, useful quality comparison can be made between a selection of encoders plus a high-quality anchor (high anchor) and low-quality anchor (low anchor) without the tester being aware of which encoder is being evaluated at any time.
Standardized quality or impairment scale
It is most common to use the 1.0 to 5.0 scale defined by ITU-R BS.1116. Any value (including fractions) between 1.0 and 5.0 is valid, with the exact whole number representing the following definitions to describe that degree of impairment:
- 5.0 : Imperceptible
- 4.0 : Perceptible, but not annoying
- 3.0 : Slightly annoying
- 2.0 : Annoying
- 1.0 : Very annoying
References
Links
This article is a stub. You can help the Hydrogenaudio Knowledgebase by expanding it.