[FFmpeg-devel,4/4] opus: Don't invert phase when downmixing to mono (per RFC8251)

Submitted by Andrew D'Addesio on Dec. 2, 2017, 5:46 p.m.

Details

Message ID 20171202174659.1064-4-modchipv12@gmail.com
State New
Headers show

Commit Message

Andrew D'Addesio Dec. 2, 2017, 5:46 p.m.
When decoding to downmixed mono, don't put the channels out of phase,
as they will cancel out and create audible artifacts. (See
RFC 8251 sec. 10.)

Signed-off-by: Andrew D'Addesio <modchipv12@gmail.com>
---
 libavcodec/opus_pvq.c | 6 ++++++
 1 file changed, 6 insertions(+)

Comments

Rostislav Pehlivanov Dec. 4, 2017, 7:39 a.m.
On 2 December 2017 at 17:46, Andrew D'Addesio <modchipv12@gmail.com> wrote:

> When decoding to downmixed mono, don't put the channels out of phase,
> as they will cancel out and create audible artifacts. (See
> RFC 8251 sec. 10.)
>
> Signed-off-by: Andrew D'Addesio <modchipv12@gmail.com>
> ---
>  libavcodec/opus_pvq.c | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/libavcodec/opus_pvq.c b/libavcodec/opus_pvq.c
> index 2f7aa74..f18c010 100644
> --- a/libavcodec/opus_pvq.c
> +++ b/libavcodec/opus_pvq.c
> @@ -643,7 +643,13 @@ static av_always_inline uint32_t
> quant_band_template(CeltPVQ *pvq, CeltFrame *f,
>                  }
>              } else {
>                  inv = (b > 2 << 3 && f->remaining2 > 2 << 3) ?
> ff_opus_rc_dec_log(rc, 2) : 0;
> +
> +                /* Don't put the channels out of phase if we are decoding
> to downmixed
> +                 * mono as this subjectively sounds bad (RFC 8251 section
> 10). */
> +                if (f->channels == 1)
> +                    inv = 0;
>              }
> +
>              itheta = 0;
>          }
>          qalloc = opus_rc_tell_frac(rc) - tell;
> --
> 2.15.1.windows.2
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>

RFC8251 made this optional, pushed a new version which added an option to
toggle this.
James Almer Dec. 4, 2017, 10:25 p.m.
On 12/4/2017 4:39 AM, Rostislav Pehlivanov wrote:
> On 2 December 2017 at 17:46, Andrew D'Addesio <modchipv12@gmail.com> wrote:
> 
>> When decoding to downmixed mono, don't put the channels out of phase,
>> as they will cancel out and create audible artifacts. (See
>> RFC 8251 sec. 10.)
>>
>> Signed-off-by: Andrew D'Addesio <modchipv12@gmail.com>
>> ---
>>  libavcodec/opus_pvq.c | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/libavcodec/opus_pvq.c b/libavcodec/opus_pvq.c
>> index 2f7aa74..f18c010 100644
>> --- a/libavcodec/opus_pvq.c
>> +++ b/libavcodec/opus_pvq.c
>> @@ -643,7 +643,13 @@ static av_always_inline uint32_t
>> quant_band_template(CeltPVQ *pvq, CeltFrame *f,
>>                  }
>>              } else {
>>                  inv = (b > 2 << 3 && f->remaining2 > 2 << 3) ?
>> ff_opus_rc_dec_log(rc, 2) : 0;
>> +
>> +                /* Don't put the channels out of phase if we are decoding
>> to downmixed
>> +                 * mono as this subjectively sounds bad (RFC 8251 section
>> 10). */
>> +                if (f->channels == 1)
>> +                    inv = 0;
>>              }
>> +
>>              itheta = 0;
>>          }
>>          qalloc = opus_rc_tell_frac(rc) - tell;
>> --
>> 2.15.1.windows.2
>>
>> _______________________________________________
>> ffmpeg-devel mailing list
>> ffmpeg-devel@ffmpeg.org
>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
> 
> RFC8251 made this optional, pushed a new version which added an option to
> toggle this.

Valgrind complains about your changes in 1b8649c2ce. Conditional jump or
move depends on uninitialised value(s).

http://fate.ffmpeg.org/report.cgi?time=20171204200723&slot=x86_64-archlinux-gcc-valgrind

Patch hide | download patch | download mbox

diff --git a/libavcodec/opus_pvq.c b/libavcodec/opus_pvq.c
index 2f7aa74..f18c010 100644
--- a/libavcodec/opus_pvq.c
+++ b/libavcodec/opus_pvq.c
@@ -643,7 +643,13 @@  static av_always_inline uint32_t quant_band_template(CeltPVQ *pvq, CeltFrame *f,
                 }
             } else {
                 inv = (b > 2 << 3 && f->remaining2 > 2 << 3) ? ff_opus_rc_dec_log(rc, 2) : 0;
+
+                /* Don't put the channels out of phase if we are decoding to downmixed
+                 * mono as this subjectively sounds bad (RFC 8251 section 10). */
+                if (f->channels == 1)
+                    inv = 0;
             }
+
             itheta = 0;
         }
         qalloc = opus_rc_tell_frac(rc) - tell;