Hello everyone, I want to use the fdk-aac library to compile an AAC+SBR audio service for drm. Does anyone have an fdk-aac library that supports 960 frame length? It would be better if there is a reference code. Thank you for your help.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You won't believe how painful it was to sign up here, just to answer this question. The captcha was being ridiculous!
The answer is 'yes'...however, it's not a slam dunk!
An OpenDigitalRadio colleague patched FDK for 960 transform so it can be used as a DAB+ encoder, however DRM30 uses the Error Robust (ER) AAC Scalable object type (ID = 20) High Quality Audio Profile and FDK doesn't have this profile.
You are right, nevertheless, I suppose a receiver would work even
without using the ER tools.
Rafael
On 3/5/24 19:42, Rash Mustapha wrote:
Hi Sam,
You won't believe how painful it was to sign up here, just to answer
this question. The captcha was being ridiculous!
The answer is 'yes'...however, it's not a slam dunk!
An OpenDigitalRadio colleague patched FDK for 960 transform so it can be
used as a DAB+ encoder, however DRM30 uses the Error Robust (ER) AAC
Scalable object type (ID = 20) High Quality Audio Profile and FDK
doesn't have this profile.
Thank you very much! Rash, I called the DAB fdk-aac library you mentioned for coding, and his code has several more parameters than the ordinary 1024 coding, I attached the screenshot of the code. But I still can't generate the aac file. In addition, after I changed the parameter TT_DABPLUS to TT_DRM, the message "Unable to set the RAW transmux" was displayed. I suspect that the parameter used by drm is different from that used by DAB. Do you know any other parameters, or do you know where to look up relevant materials。Thank you again！
I don't know enough to be certain that it would, but If someone does plug FDK in, I would be happy to carry out tests against a DRM codec that does have ER. I don't have a suitable channel simulator, so only 'conducted' measurements to several different receivers (e.g. a Coding Technologies portable) via a variable step attenuator, but it should be adequate for quantifying how valuable ER is.
R
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You can also look at the ODR-AudioEnc source to see how FDK-AAC is interfaced for DAB+.
All of the DRM related bits have been stripped out of FDK-AAC, which is probably the source of your error?
So I suspect you'd need to call it as DAB+, or better still, populate the missing DRM bits with the DAB+ code and hack away on that.
I don't know where/when the superframing happens (i.e. inside or outside of FDK-AAC). If it's inside, that's a major difference because DRM uses 400mS superframes (Modes A to D) or 200mS (mode E) and DAB+ is 120mS.
Hopefully there's something in the above which might be useful!
R
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It supports many object types, including HE-AAC and USAC (xHE-AAC). USAC
is also supported in DRM (thou the super-framing is a bit different in
order to allow "abr" framing).
I can see 960 frames support is there, and also some ER tools!
Cheers,
Rafael
On 3/7/24 13:41, Rash wrote:
Hi Sam,
Apologies - I don't really know enough to be much help.
You should probably look at this repo instead of the one I previously
pointed you at, because I think the fdk included is much more up to date:
You can also look at the ODR-AudioEnc source to see how FDK-AAC is
interfaced for DAB+.
All of the DRM related bits have been stripped out of FDK-AAC, which is
probably the source of your error?
So I suspect you'd need to call it as DAB+, or better still, populate
the missing DRM bits with the DAB+ code and hack away on that.
I don't know where/when the superframing happens (i.e. inside or outside
of FDK-AAC). If it's inside, that's a major difference because DRM uses
400mS superframes (Modes A to D) or 200mS (mode E) and DAB+ is 120mS.
Hopefully there's something in the above which might be useful!
It supports many object types, including HE-AAC and USAC (xHE-AAC). USAC
is also supported in DRM (thou the super-framing is a bit different in
order to allow "abr" framing).
I can see 960 frames support is there, and also some ER tools!
Cheers,
Rafael
On 3/7/24 13:41, Rash wrote:
Hi Sam,
Apologies - I don't really know enough to be much help.
You should probably look at this repo instead of the one I previously
pointed you at, because I think the fdk included is much more up to date:
You can also look at the ODR-AudioEnc source to see how FDK-AAC is
interfaced for DAB+.
All of the DRM related bits have been stripped out of FDK-AAC, which is
probably the source of your error?
So I suspect you'd need to call it as DAB+, or better still, populate
the missing DRM bits with the DAB+ code and hack away on that.
I don't know where/when the superframing happens (i.e. inside or outside
of FDK-AAC). If it's inside, that's a major difference because DRM uses
400mS superframes (Modes A to D) or 200mS (mode E) and DAB+ is 120mS.
Hopefully there's something in the above which might be useful!
Hello everyone, I want to use the fdk-aac library to compile an AAC+SBR audio service for drm. Does anyone have an fdk-aac library that supports 960 frame length? It would be better if there is a reference code. Thank you for your help.
Hi Sam,
You won't believe how painful it was to sign up here, just to answer this question. The captcha was being ridiculous!
The answer is 'yes'...however, it's not a slam dunk!
An OpenDigitalRadio colleague patched FDK for 960 transform so it can be used as a DAB+ encoder, however DRM30 uses the Error Robust (ER) AAC Scalable object type (ID = 20) High Quality Audio Profile and FDK doesn't have this profile.
It's here anyway: https://github.com/Opendigitalradio/fdk-aac
And good luck!
R
You are right, nevertheless, I suppose a receiver would work even
without using the ER tools.
On 3/5/24 19:42, Rash Mustapha wrote:
Thank you very much! Rash, I called the DAB fdk-aac library you mentioned for coding, and his code has several more parameters than the ordinary 1024 coding, I attached the screenshot of the code. But I still can't generate the aac file. In addition, after I changed the parameter TT_DABPLUS to TT_DRM, the message "Unable to set the RAW transmux" was displayed. I suspect that the parameter used by drm is different from that used by DAB. Do you know any other parameters, or do you know where to look up relevant materials。Thank you again！
Hi Rafael,
That's a good point!
I don't know enough to be certain that it would, but If someone does plug FDK in, I would be happy to carry out tests against a DRM codec that does have ER. I don't have a suitable channel simulator, so only 'conducted' measurements to several different receivers (e.g. a Coding Technologies portable) via a variable step attenuator, but it should be adequate for quantifying how valuable ER is.
R
Hi Sam,
Apologies - I don't really know enough to be much help.
You should probably look at this repo instead of the one I previously pointed you at, because I think the fdk included is much more up to date:
https://github.com/Opendigitalradio/ODR-AudioEnc/tree/master
You can also look at the ODR-AudioEnc source to see how FDK-AAC is interfaced for DAB+.
All of the DRM related bits have been stripped out of FDK-AAC, which is probably the source of your error?
So I suspect you'd need to call it as DAB+, or better still, populate the missing DRM bits with the DAB+ code and hack away on that.
I don't know where/when the superframing happens (i.e. inside or outside of FDK-AAC). If it's inside, that's a major difference because DRM uses 400mS superframes (Modes A to D) or 200mS (mode E) and DAB+ is 120mS.
Hopefully there's something in the above which might be useful!
R
I would also consider libxaac, which explicitly mentions DRM:
https://github.com/ittiam-systems/libxaac
It supports many object types, including HE-AAC and USAC (xHE-AAC). USAC
is also supported in DRM (thou the super-framing is a bit different in
order to allow "abr" framing).
I can see 960 frames support is there, and also some ER tools!
Cheers,
Rafael
On 3/7/24 13:41, Rash wrote:
Hello,
I wasn't aware of this development - this is very interesting indeed!
It's probably better to focus on this encoder rather than FDK-AAC.
R
On Thu, 7 Mar 2024, 16:25 Rafael Diniz, rafael2k@users.sourceforge.net
wrote: