Commit Graph

40 Commits

Author SHA1 Message Date
903182bd8e Revert r8476 "Set decoder output frequency in AudioDecoder::Decode call"
This change uncovered issue 4143, evading the Memcheck suppression
since the signature is changed in the Decode function.

A fix for this is in the making; see
https://review.webrtc.org/36309004. This CL will be re-landed once the
fix is in place.

BUG=4143
TBR=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/42089004

Cr-Commit-Position: refs/heads/master@{#8488}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8488 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-02-24 21:18:44 +00:00
b9c18d5643 Set decoder output frequency in AudioDecoder::Decode call
This CL changes the way the decoder sample rate is set and updated. In
practice, it only concerns the iSAC (float) codec.

One single iSAC decoder instance is used for both wideband and
super-wideband decoding, and the instance must be told to switch
output frequency if the payload type changes. This used to be done
through a call to UpdateDecoderSampleRate, but is now instead done in
the Decode call as an extra parameter.

R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/34349004

Cr-Commit-Position: refs/heads/master@{#8476}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8476 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-02-24 15:59:20 +00:00
d324546ced Misc. cleanup split out of https://webrtc-codereview.appspot.com/37699004/ :
* Move constants into the files/functions that use them
* Declare variables in the narrowest scope possible
* Use correct (expected, actual) order for gtest macros
* Remove unused functions
* Untabify
* 80-column limit
* Avoid C-style casts
* Prefer true typed constants to "enum hack" constants
* Print size_t using the right format macro
* Shorten and simplify code
* Other random cleanup bits and style fixes

BUG=none
TEST=none
R=henrik.lundin@webrtc.org, tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/36179004

Cr-Commit-Position: refs/heads/master@{#8467}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8467 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-02-23 21:29:45 +00:00
c0bd7be0df Adding two new stats to VoiceReceiverInfo
There have been requests of two new stats namely

speech_expand_rate and secondary_decoded_rate.

BUG=3867
R=henrik.lundin@webrtc.org, henrika@webrtc.org, tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/40789004

Cr-Commit-Position: refs/heads/master@{#8415}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8415 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-02-18 15:24:39 +00:00
026b892e72 Using << on an int8_t or uint8_t will output a character rather than a number.
Places that do this need to cast to int to get the desired behavior.

BUG=none
TEST=none
R=henrik.lundin@webrtc.org, pthatcher@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/40579004

Cr-Commit-Position: refs/heads/master@{#8223}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8223 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-01-30 19:54:19 +00:00
16825b1a82 Use int64_t more consistently for times, in particular for RTT values.
Existing code was inconsistent about whether to use uint16_t, int, unsigned int,
or uint32_t, and sometimes silently truncated one to another, or truncated
int64_t.  Because most core time-handling functions use int64_t, being
consistent about using int64_t unless otherwise necessary minimizes the number
of explicit or implicit casts.

BUG=chromium:81439
TEST=none
R=henrik.lundin@webrtc.org, holmer@google.com, tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/31349004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@8045 4adac7df-926f-26a2-2b94-8c16560cd09d
2015-01-12 21:51:21 +00:00
e04a93bcf5 Move the AudioDecoder interface out of NetEq
It belongs with the codecs, next to the AudioEncoder interface.

R=andrew@webrtc.org, henrik.lundin@webrtc.org, kjellander@webrtc.org

Previously committed here: https://code.google.com/p/webrtc/source/detail?r=7798
and reverted here: https://code.google.com/p/webrtc/source/detail?r=7799

Review URL: https://webrtc-codereview.appspot.com/27309004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7839 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-12-09 10:12:53 +00:00
3800e13a3a Revert r7798 ("Move the AudioDecoder interface out of NetEq")
Apparently, it caused all sorts of problems I don't have time to
straighten out right now.

TBR=henrika@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25289004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7799 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-12-03 16:28:17 +00:00
00ba1a7dfd Move the AudioDecoder interface out of NetEq
It belongs with the codecs, next to the AudioEncoder interface.

R=henrik.lundin@webrtc.org, kjellander@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27309004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7798 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-12-03 14:23:23 +00:00
4591fbd09f Use size_t more consistently for packet/payload lengths.
See design doc at https://docs.google.com/a/chromium.org/document/d/1I6nmE9D_BmCY-IoV6MDPY2V6WYpEI-dg2apWXTfZyUI/edit?usp=sharing for more information.

This CL was reviewed and approved in pieces in the following CLs:
https://webrtc-codereview.appspot.com/24209004/
https://webrtc-codereview.appspot.com/24229004/
https://webrtc-codereview.appspot.com/24259004/
https://webrtc-codereview.appspot.com/25109004/
https://webrtc-codereview.appspot.com/26099004/
https://webrtc-codereview.appspot.com/27069004/
https://webrtc-codereview.appspot.com/27969004/
https://webrtc-codereview.appspot.com/27989004/
https://webrtc-codereview.appspot.com/29009004/
https://webrtc-codereview.appspot.com/30929004/
https://webrtc-codereview.appspot.com/30939004/
https://webrtc-codereview.appspot.com/31999004/
Committing as TBR to the original reviewers.

BUG=chromium:81439
TEST=none
TBR=pthatcher,henrik.lundin,tina.legrand,stefan,tkchin,glaznev,kjellander,perkj,mflodman,henrika,asapersson,niklas.enbom

Review URL: https://webrtc-codereview.appspot.com/23129004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7726 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-11-20 22:28:14 +00:00
913f7b8d5e Fix for glitches in ACM when switching desired output sample rate
The problem was that if the output sample rate is changed such from one
where no resampling is needed to a rate that requires resampling, the
first output from the resampler will contain an onset period. The
solution provided in this CL is to keep a copy of the last output frame
in ACM, and if the resampler is engaged, it will be primed with this
old frame before resampling the current frame.

BUG=3919
R=bjornv@webrtc.org, turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27729004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7479 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-10-21 06:54:23 +00:00
ea25784107 Change how background noise mode in NetEq is set
This change prepares for switching default background noise (bgn) mode
from on to off. The actual switch will be done later.

In this change, the bgn mode is included as a setting in NetEq's config
struct. We're also removing the connection between playout modes and
bgn modes in ACM. In practice this means that bgn mode will change from
off to on for streaming mode, but since the playout modes are not used
it does not matter.

BUG=3519
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21749004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6843 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-08-07 12:27:37 +00:00
eecf5e6ba7 Removing neteq decode lock and friends
NetEq is thread-safe by virtue of it's own lock, and in r6404 the
ACMISAC class was made thread-safe. Therefore, the neteq decode lock
is no longer needed.

R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/18599004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6531 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-06-24 13:11:22 +00:00
a90abdef62 Add thread annotations to AcmReceiver
This change adds thread annotations to AcmReceiver. These are the
annotations that could be added without changing acquiring the locks in
more locations, or changing the lock structure.

BUG=3401
R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/17649004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6376 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-06-09 18:35:11 +00:00
9c55f0f957 Rename neteq4 folder to neteq
Keep the old neteq4/audio_decoder_unittests.isolate while waiting for
a hard-coded reference to change.

This CL effectively reverts r6257 "Rename neteq4 folder to neteq".

BUG=2996
TBR=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/21629004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6367 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-06-09 08:10:28 +00:00
94454b71ad Fix the chain that propagates the audio frame's rtp and ntp timestamp including:
* In AudioCodingModuleImpl::PlayoutData10Ms, don't reset the timestamp got from GetAudio.
* When there're more than one participant, set AudioFrame's RTP timestamp to 0.
* Copy ntp_time_ms_ in AudioFrame::CopyFrom method.
* In RemixAndResample, pass src frame's timestamp_ and ntp_time_ms_ to the dst frame.
* Fix how |elapsed_time_ms| is computed in channel.cc by adding GetPlayoutFrequency.

Tweaks on ntp_time_ms_:
* Init ntp_time_ms_ to -1 in AudioFrame ctor.
* When there're more than one participant, set AudioFrame's ntp_time_ms_ to an invalid value. I.e. we don't support ntp_time_ms_ in multiple participants case before the mixing is moved to chrome.

Added elapsed_time_ms to AudioFrame and pass it to chrome, where we don't have the information about the rtp timestmp's sample rate, i.e. can't convert rtp timestamp to ms.

BUG=3111
R=henrik.lundin@webrtc.org, turaj@webrtc.org, xians@webrtc.org
TBR=andrew
andrew to take another look on audio_conference_mixer_impl.cc

Review URL: https://webrtc-codereview.appspot.com/14559004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6346 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-06-05 20:34:08 +00:00
1b9df05c85 Revert 6257 "Rename neteq4 folder to neteq"
> Rename neteq4 folder to neteq
> 
> BUG=2996
> R=turaj@webrtc.org
> 
> Review URL: https://webrtc-codereview.appspot.com/12569005

TBR=henrik.lundin@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13549004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6259 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-28 07:33:39 +00:00
a90f6d67f7 Rename neteq4 folder to neteq
BUG=2996
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12569005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6257 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-28 06:23:34 +00:00
cb711f77d2 Add interface to propagate audio capture timestamp to the renderer.
BUG=3111
R=andrew@webrtc.org, turaj@webrtc.org, xians@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12239004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6189 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-19 17:39:11 +00:00
5c49c64de5 Remove all use of AudioFrame::energy_ from AudioCodingModule
Since r6117, the energy is always calculated in the mixer module,
regardless of the value that ACM sets for energy_.

This part of the the aftermath of issue 3255.

BUG=3255
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/19459004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6140 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-05-14 09:06:52 +00:00
0bc9b5a5a7 Add clock to ACM config struct
The purpose is to clean up the ACM interface a bit. This is a
follow-up of a comment in http://review.webrtc.org/13379004/.

R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/16389005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@6006 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-29 08:09:31 +00:00
e772c71743 Introduce a config struct for AudioCoding module
The config struct currently contains the module ID, and the NetEq
config struct, but will be extended in the future. The purpose of this
change is to expose certain NetEq settings to the ACM interface.

BUG=3083
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/13379004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5993 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-28 10:16:57 +00:00
116ed1d4f0 Include buffer size limits in NetEq config struct
This change includes max_packets_in_buffer and max_delay_ms in the
NetEq config struct. The packet buffer is also no longer limited in
terms of payload sizes (bytes), only number of packets.

The old constants governing the packet buffer limits are deleted.

BUG=3083
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14389004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5989 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-28 08:20:04 +00:00
439a4c49f9 Add an output capacity parameter to ACMResampler::Resample10Msec()
Also adding a unit tests to make sure that a desired output frequency
of 0 passed to AudioCodingModule::PlayoutData10Ms() is invalid.

R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/14369005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5974 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-24 19:05:33 +00:00
20c71fd1dc Fix a bug in AcmReceiver::NetworkStatistics
One of the variables were not copied between the structs.

BUG=2996
TBR=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12329004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5956 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-22 10:11:21 +00:00
0c1444c748 Create ACM2 instance when calling AudioCodingModule::Create
BUG=2996
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12079005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5952 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-22 08:18:42 +00:00
a596a389ea Fix iSAC/48000 issue with ACM2.
Registeration of iSAC into NetEq is through injecting and external AudioDecoder. This is because iSAC encoder and decoder need to share instances for bandwidth estimator to work. When external decoder is registerred, the sampling rate of the decoder had to be specified. iSAC/48000 decoder has a native sampling rate of 32000 Hz, but it has been registered as 48000 Hz decoder.

This CL fixing this issue by letting NetEq to obtain sampling rate of an external coder according to its existing database.

BUG=3143
TEST=voe_cmd_test,modules_unittest,try-bots
R=henrik.lundin@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/12139004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5936 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-17 23:30:49 +00:00
35ead381f8 Adding a config struct to NetEq
With this change, the parameters sent to the NetEq::Create method are
collected in one NetEq::Config struct. The benefit is that it is easier
to set, change and override default values, and easier to expand with
more parameters in the future.

BUG=3083
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/11949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5902 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-04-14 18:49:17 +00:00
ba5a6c3d89 ACM2/NetEq4 did not decode Opus in stereo
Two problems fixed in this CL:
- setting Opus decoder to stereo had no effect, and decoding always generated mono audio
- changing decoding setting from mono to stereo, or stereo to mono, for OPUS also had no effect (but required another change than the first one).

BUG=3082
R=henrik.lundin@webrtc.org, turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/10389004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5754 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-03-23 09:58:48 +00:00
2086e0fbf3 Remove unnecessary warnings.
BUG=
TEST=try job
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/8719005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5565 4adac7df-926f-26a2-2b94-8c16560cd09d
2014-02-18 14:22:20 +00:00
24301a67c6 Update talk to 58174641 together with http://review.webrtc.org/4319005/.
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5809004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5287 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-12-13 19:17:43 +00:00
a92baead39 ACM 2 compatibility with ACM 1.
Removing an unregisterd codec from ACM 1 does not result in an error, so should be for ACM 2. Also ACM 1 has post-decode VAD on and AMC 2 needs to have it on by default.

BUG=
Test=trybits

R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5549005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5276 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-12-13 00:10:44 +00:00
6d5d248075 Wrap ACM2 code inside acm2 namespace. This gurantees that one ACM would not use components of others by accident.
BUG=
R=minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2344004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4933 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-10-06 04:47:28 +00:00
3e7703640f Remove unused constants, so chrome can enable a warning for that. Patch from thakis@
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2296006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4844 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-25 22:05:05 +00:00
d6a7a5f385 Small fixes to run ACM2 tests.
BUG=
R=minyue@google.com

Review URL: https://webrtc-codereview.appspot.com/2238004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4836 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-25 01:09:23 +00:00
532f3dc548 Compile ACM2 and ACM1.
First patch set is the same as patch set 3 of http://review.webrtc.org/2237004/

-Make ACM1 to depend on ACM2.
-Remove APIs to set and get background noise mode. There is no VoE call to these
APIs.
-Remove APIs to set and get receive side VAD mode. There is no VoE call to these
APIs, and NetEq 4, doesn't support them.
-Remove callback for in-band DTMF detection. ACM doesn't support in-band DTMF
detection.
-Use acm_common_defs.h everywhere required.
-Complete ACM factory method.
-Update ACMCodecDatabase of ACM2. CNG full-band need to be define-guarded.
Remove dynamic payload-type assignment.

BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2249004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4785 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-19 00:12:23 +00:00
1c77dfd521 Revert r4772 "Compile ACM1 and ACM2."
Breaks Android build.

TBR=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2244004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4777 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-18 12:34:05 +00:00
367baa6eb3 Compile ACM1 and ACM2.
-Make ACM1 to depend on ACM2.
-Remove APIs to set and get background noise mode. There is no VoE call to these APIs.
-Remove APIs to set and get receive side VAD mode. There is no VoE call to these APIs, and NetEq 4, doesn't support them.
-Remove callback for in-band DTMF detection. ACM doesn't support in-band DTMF detection.
-Use acm_common_defs.h everywhere required.
-Complete ACM factory method.
-Update ACMCodecDatabase of ACM2. CNG full-band need to be define-guarded. Remove dynamic payload-type assignment.

BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2237004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4772 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-18 00:36:11 +00:00
48af652ea5 Prepare to compile ACM1 and ACM2.
ACM1 code is wrapped in namespace acm1. Inculde paths and define guards of ACM2 source codes are corrected. gypi file of ACM2 is changed so that ACM1 will later on depends on ACM2.

BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2206004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4743 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-13 23:06:59 +00:00
7959e16cc2 ACM2 integration with NetEq 4.
nack{.cc, .h, _unittest.cc} are basically copies from main/source/ folder, with cpplint warning cleaned up.

BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2190009

git-svn-id: http://webrtc.googlecode.com/svn/trunk@4736 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-09-12 18:30:26 +00:00