shithub: opus

Info  •  Files  •  Log  •  Branches

draft: mode switching details (reset and redundancy cross-fade)

--- a/doc/draft-ietf-codec-opus.xml

+++ b/doc/draft-ietf-codec-opus.xml

@@ -5293,27 +5293,34 @@

 </t>

 <t>

-For CELT-only to SILK-only or hybrid transitions, the first 2.5 ms of the

- redundant frame is used as-is for the reconstructed output.

-The remaining 2.5 ms is overlapped and added (cross-faded using the square

+If the redundancy belongs at the beginning (CELT-only to SILK-only or Hybrid transitions),

+the first 2.5 ms of the redundant frame is used as-is for the first 2.5 ms of the reconstructed output.

+The remaining 2.5 ms is overlapped and added (faded out using the square

  of the MDCT power-complementary window) to the decoded SILK/hybrid signal,

  ensuring a smooth transition.

-For SILK-only or hyrid to CELT-only transitions, only the second half of the

- redundant frame is used.

-In that case, only a 2.5 ms cross-fade is applied, still using the

- power-complementary window.

-<!--TODO: I don't understand this at all.

-    A 5 ms frame with the CELT window applied applied has 7.5 ms of output:

-     2.5 ms of fade-in, 2.5 ms unwindowed, and 2.5 ms of fade-out.

-    Which portions are being referred to above?

-    How are they aligned with the rest of the stream?

-

-    Also, the bitstream can include redundancy on other transitions than the

-     ones listed in this paragraph.

-    What's the required behavior?-->

+If the redundancy belongs at the end (SILK-only or hyrid to CELT-only transitions), 

+only the second half (2.5 ms) of the redundant frame is used.

+In that case, the second half of the redundant frame is faded in using a 2.5 ms cross-fade applied

+at the end of the reconstructed output. This also uses the power-complementary window.

 </t>

 </section>

+</section>

+

+<section anchor="decoder-reset" title="State Reset">

+<t>When some transitions occur, the state of the SILK or the CELT decoder (or both)

+needs to be reset before decoding a frame in the new mode, in part to avoid reusing

+"out of date" memory. The SILK state is

+restarted every time we decode a SILK-only or Hybrid frame and the previous frame 

+was CELT-only. For the CELT state, the general rule is that it is restarted every time

+we switch between the three modes and the new mode is either Hybrid or CELT-only. The

+exception to this rule is when transition side information is used. When switching from

+SILK-only or Hybrid to CELT-only mode with redundancy, then the CELT state is reset 

+before decoding the redundant CELT frame embedded in the SILK-only/Hybrid frame, but it is not

+before decoding the following CELT-only frame. When switching from CELT-only mode to SILK-only

+or Hybrid mode with redundancy, the CELT decoder is not reset for decoding the CELT

+redundant frame.

+</t>

 </section>

 </section>