Diego Alejandro Parra Guzman
2017-Apr-11 09:04 UTC
[opus] MDCT implementation and his overlapped relationship
Dear all I'm working on the implementation of the MDCT for the processor ARM-Cortex-M4 I'm trying to replicate the behavior of the MDCT for several overlapped values however I realized that current implementation of the MDCT is very close to the theory only in case in which we have and overlap exactly equal to N/2 where N is the size of the input vector as is shown in the examples for celt. could you explain me if the implementation had been thought for different overlapped values, or instead the implementation uses only the overlap equal to N/2. Additional I want to know in case of overlapped values different to N/2 which values had been considered. Thanks for you support. -- Diego Alejandro Parra Guzmán Estudiante de ingeniería electrónica Universidad distrital FJC -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.xiph.org/pipermail/opus/attachments/20170411/b3ef3c3f/attachment-0001.html>
Jean-Marc Valin
2017-Apr-13 06:59 UTC
[opus] MDCT implementation and his overlapped relationship
Hi Diego, You might find figure 3 or this paper useful: https://jmvalin.ca/papers/aes135_opus_celt.pdf It shows the MDCT window for a 5 ms frame, but the (non-zero) overlap is always 120 samples regarless of the frame size. You can either see it as partial overlap, or as a full-overlap with a window that has zeros on the sides. Cheers, Jean-Marc On 11/04/17 05:04 AM, Diego Alejandro Parra Guzman wrote:> Dear all > > I'm working on the implementation of the MDCT for the processor > ARM-Cortex-M4 I'm trying to replicate the behavior of the MDCT > for several overlapped values however I realized that current > implementation of the MDCT is very close to the theory only in case in > which we have and overlap exactly equal to N/2 where N is the size of > the input vector as is shown in the examples for celt. > > could you explain me if the implementation had been thought for > different overlapped values, or instead the implementation uses only the > overlap equal to N/2. > > Additional I want to know in case of overlapped values different to N/2 > which values had been considered. > > Thanks for you support. > > -- > Diego Alejandro Parra Guzmán > Estudiante de ingeniería electrónica > Universidad distrital FJC > > > _______________________________________________ > opus mailing list > opus at xiph.org > http://lists.xiph.org/mailman/listinfo/opus >
Diego Alejandro Parra Guzman
2017-May-03 09:01 UTC
[opus] MDCT implementation and his overlapped relationship
Dear all In the paper section describes how CELT and SILK had implemented a look/up head delay of (2.5ms), (5ms) respectively , could you explain me> 1) those values are fixed. 1) how you guarantee those values independent of the architecture's performance. Thanks so much On Tue, Apr 18, 2017 at 10:05 AM, Diego Alejandro Parra Guzman < daparrag at correo.udistrital.edu.co> wrote:> thanks you so much for your answer. > > On Thu, Apr 13, 2017 at 8:59 AM, Jean-Marc Valin <jmvalin at jmvalin.ca> > wrote: > >> Hi Diego, >> >> You might find figure 3 or this paper useful: >> https://jmvalin.ca/papers/aes135_opus_celt.pdf >> >> It shows the MDCT window for a 5 ms frame, but the (non-zero) overlap is >> always 120 samples regarless of the frame size. You can either see it as >> partial overlap, or as a full-overlap with a window that has zeros on >> the sides. >> >> Cheers, >> >> Jean-Marc >> >> On 11/04/17 05:04 AM, Diego Alejandro Parra Guzman wrote: >> > Dear all >> > >> > I'm working on the implementation of the MDCT for the processor >> > ARM-Cortex-M4 I'm trying to replicate the behavior of the MDCT >> > for several overlapped values however I realized that current >> > implementation of the MDCT is very close to the theory only in case in >> > which we have and overlap exactly equal to N/2 where N is the size of >> > the input vector as is shown in the examples for celt. >> > >> > could you explain me if the implementation had been thought for >> > different overlapped values, or instead the implementation uses only the >> > overlap equal to N/2. >> > >> > Additional I want to know in case of overlapped values different to N/2 >> > which values had been considered. >> > >> > Thanks for you support. >> > >> > -- >> > Diego Alejandro Parra Guzmán >> > Estudiante de ingeniería electrónica >> > Universidad distrital FJC >> > >> > >> > _______________________________________________ >> > opus mailing list >> > opus at xiph.org >> > http://lists.xiph.org/mailman/listinfo/opus >> > >> > > > > -- > Diego Alejandro Parra Guzmán > Estudiante de ingeniería electrónica > Universidad distrital FJC >-- Diego Alejandro Parra Guzmán Estudiante de ingeniería electrónica Universidad distrital FJC -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.xiph.org/pipermail/opus/attachments/20170503/dd41315a/attachment.html>