Purité Audio
Trade: Purite Audio
Looks interesting,
http://www.dirac.se/en/news-events/...irely-new-approach-to-sound-optimization.aspx
Keith
http://www.dirac.se/en/news-events/...irely-new-approach-to-sound-optimization.aspx
Keith
Dirac Unison released
- Thread starter Purité Audio
- Start date
Purité Audio
Trade: Purite Audio
That's the story of my life!
AES papaer from this year, http://www.aes.org/e-lib/browse.cfm?elib=17563
Kr Keith.
AES papaer from this year, http://www.aes.org/e-lib/browse.cfm?elib=17563
Kr Keith.
Purité Audio
Trade: Purite Audio
Extracts from paper,
With reference to Dirac's Unison a very recent paper has been published by the Audio Engineering Society... it is dated February 10, 2015 and is available in full here:
http://www.aes.org/e-lib/browse.cfm?elib=17563
The following is an excerpt with relevant references:
"To obtain a specified target response in several measurement positions is a difficult task, particularly with individually designed filters.
First, as nearly all loudspeaker–room impulse responses exhibit excess phase (non-minimum phase) behavior [17, 25] the pre-compensator must be of so-called mixed phase type to correct for the excess phase distortion, i.e., the distortion components that are non-minimum phase.
Minimum phase filters, which are most commonly used, are generally insufficient for correcting such components.
Second, if a loudspeaker impulse response varies significantly between different measurement positions, as is typically the case in a normal room, then a single individual filter design for each speaker would, in general, not be sufficient to obtain good performance in all measurement positions.
A response may be attained that on average is close to the specifications, but there will always be remaining errors at each measurement position.
Hence, single-channel methods are most effective for compensating distortions that are common, or nearly common, to all positions in the region of interest.
Since audio systems of today generally include multiple loudspeaker channels, modern room correction methods propose the use of all, or at least a subset of, the available loudspeakers, see, e.g., [1, 2, 9, 12, 13, 18, 21, 23, 30–32].
In a recent publication, a Multiple-Input Multiple-Output(MIMO) approach to room compensation by the use of support loudspeakers was proposed and evaluated for a varying number of support loudspeakers [9].
It was shown that the effect of the room acoustics can be completely controlled up to a certain frequency, which is determined by factors such as, e.g., the number of loudspeakers, the size of the room and the sweet spot, and the granularity of the grid of measurement positions.
In the present paper we shall investigate how the MIMO framework can be used to variably control the contribution of the listening room acoustics throughout a spatial region where the listener is located.
By allowing the support loudspeakers to help the primary loudspeakers (e.g., the left and right speakers in a stereo system) to a higher or lesser degree, we can obtain a range of different equalizers that, to corresponding degrees, suppress the room acoustics while the direct sound of the primary loudspeakers is enhanced.
Consequently we are able to decrease the influence of the room in all spatial positions simultaneously, to an extent that can be determined by the user.
In the sequel we will refer to this concept as “Focus Control.”
The motivation for this type of variable room equalization can be manifold and depends on the role of the listener, on the audio material, and on the practical situation.
For example, a professional listener such as a mixing or mastering engineer can use it as a tool for carefully examining how various levels of listening room acoustics affect the perception of the material being produced.
At the consumer end, the preferences and needs of listeners may vary depending on the material being listened to.
If maximum intelligibility is desired for, e.g., a speech recording, then the maximum focus level (minimum level of listening room acoustics) is most likely preferable.
On the other hand, if the recording is a musical performance in a studio, involving a voice or an acoustic solo instrument, then it may be better not to remove too much of the listening room’s response, since the performance may then appear overly dry or dull, lacking the natural acoustic space normally associated with such performances"
With reference to Dirac's Unison a very recent paper has been published by the Audio Engineering Society... it is dated February 10, 2015 and is available in full here:
http://www.aes.org/e-lib/browse.cfm?elib=17563
The following is an excerpt with relevant references:
"To obtain a specified target response in several measurement positions is a difficult task, particularly with individually designed filters.
First, as nearly all loudspeaker–room impulse responses exhibit excess phase (non-minimum phase) behavior [17, 25] the pre-compensator must be of so-called mixed phase type to correct for the excess phase distortion, i.e., the distortion components that are non-minimum phase.
Minimum phase filters, which are most commonly used, are generally insufficient for correcting such components.
Second, if a loudspeaker impulse response varies significantly between different measurement positions, as is typically the case in a normal room, then a single individual filter design for each speaker would, in general, not be sufficient to obtain good performance in all measurement positions.
A response may be attained that on average is close to the specifications, but there will always be remaining errors at each measurement position.
Hence, single-channel methods are most effective for compensating distortions that are common, or nearly common, to all positions in the region of interest.
Since audio systems of today generally include multiple loudspeaker channels, modern room correction methods propose the use of all, or at least a subset of, the available loudspeakers, see, e.g., [1, 2, 9, 12, 13, 18, 21, 23, 30–32].
In a recent publication, a Multiple-Input Multiple-Output(MIMO) approach to room compensation by the use of support loudspeakers was proposed and evaluated for a varying number of support loudspeakers [9].
It was shown that the effect of the room acoustics can be completely controlled up to a certain frequency, which is determined by factors such as, e.g., the number of loudspeakers, the size of the room and the sweet spot, and the granularity of the grid of measurement positions.
In the present paper we shall investigate how the MIMO framework can be used to variably control the contribution of the listening room acoustics throughout a spatial region where the listener is located.
By allowing the support loudspeakers to help the primary loudspeakers (e.g., the left and right speakers in a stereo system) to a higher or lesser degree, we can obtain a range of different equalizers that, to corresponding degrees, suppress the room acoustics while the direct sound of the primary loudspeakers is enhanced.
Consequently we are able to decrease the influence of the room in all spatial positions simultaneously, to an extent that can be determined by the user.
In the sequel we will refer to this concept as “Focus Control.”
The motivation for this type of variable room equalization can be manifold and depends on the role of the listener, on the audio material, and on the practical situation.
For example, a professional listener such as a mixing or mastering engineer can use it as a tool for carefully examining how various levels of listening room acoustics affect the perception of the material being produced.
At the consumer end, the preferences and needs of listeners may vary depending on the material being listened to.
If maximum intelligibility is desired for, e.g., a speech recording, then the maximum focus level (minimum level of listening room acoustics) is most likely preferable.
On the other hand, if the recording is a musical performance in a studio, involving a voice or an acoustic solo instrument, then it may be better not to remove too much of the listening room’s response, since the performance may then appear overly dry or dull, lacking the natural acoustic space normally associated with such performances"