zqz/platform-external-webrtc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial upload of NetEq4

Browse Source 
This is the first public upload of the new NetEq, version 4.

It has been through extensive internal review during the course of
the project.

TEST=trybots

Review URL: https://webrtc-codereview.appspot.com/1073005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3425 4adac7df-926f-26a2-2b94-8c16560cd09d
This commit is contained in:
henrik.lundin@webrtc.org
2013-01-29 12:09:21 +00:00
parent 63e0964039
commit d94659dc27
129 changed files with 24101 additions and 103 deletions
Download Patch File Download Diff File Expand all files Collapse all files

227
webrtc/modules/audio_coding/neteq4/interface/neteq.h Normal file
View File

@ -0,0 +1,227 @@
/*
* Copyright (c) 2012 The WebRTC project authors. All Rights Reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#ifndef WEBRTC_MODULES_AUDIO_CODING_NETEQ4_INTERFACE_NETEQ_H_
#define WEBRTC_MODULES_AUDIO_CODING_NETEQ4_INTERFACE_NETEQ_H_
#include <cstring> // Provide access to size_t.
#include <vector>
#include "webrtc/modules/audio_coding/neteq4/interface/audio_decoder.h"
#include "webrtc/system_wrappers/interface/constructor_magic.h"
#include "webrtc/typedefs.h"
namespace webrtc {
// Forward declarations.
struct WebRtcRTPHeader;
// RTCP statistics.
struct RtcpStatistics {
uint16_t fraction_lost;
uint32_t cumulative_lost;
uint32_t extended_max;
uint32_t jitter;
};
struct NetEqNetworkStatistics {
uint16_t current_buffer_size_ms; // Current jitter buffer size in ms.
uint16_t preferred_buffer_size_ms; // Target buffer size in ms.
uint16_t jitter_peaks_found; // 1 if adding extra delay due to peaky
// jitter; 0 otherwise.
uint16_t packet_loss_rate; // Loss rate (network + late) in Q14.
uint16_t packet_discard_rate; // Late loss rate in Q14.
uint16_t expand_rate; // Fraction (of original stream) of synthesized
// speech inserted through expansion (in Q14).
uint16_t preemptive_rate; // Fraction of data inserted through pre-emptive
// expansion (in Q14).
uint16_t accelerate_rate; // Fraction of data removed through acceleration
// (in Q14).
int32_t clockdrift_ppm; // Average clock-drift in parts-per-million
// (positive or negative).
int added_zero_samples; // Number of zero samples added in "off" mode.
};
enum NetEqOutputType {
kOutputNormal,
kOutputPLC,
kOutputCNG,
kOutputPLCtoCNG,
kOutputVADPassive
};
enum NetEqPlayoutMode {
kPlayoutOn,
kPlayoutOff,
kPlayoutFax,
kPlayoutStreaming
};
// This is the interface class for NetEq.
class NetEq {
public:
enum ReturnCodes {
kOK = 0,
kFail = -1,
kNotImplemented = -2
};
enum ErrorCodes {
kNoError = 0,
kOtherError,
kInvalidRtpPayloadType,
kUnknownRtpPayloadType,
kCodecNotSupported,
kDecoderExists,
kDecoderNotFound,
kInvalidSampleRate,
kInvalidPointer,
kAccelerateError,
kPreemptiveExpandError,
kComfortNoiseErrorCode,
kDecoderErrorCode,
kOtherDecoderError,
kInvalidOperation,
kDtmfParameterError,
kDtmfParsingError,
kDtmfInsertError,
kStereoNotSupported,
kSampleUnderrun,
kDecodedTooMuch,
kFrameSplitError,
kRedundancySplitError,
kPacketBufferCorruption
};
static const int kMaxNumPacketsInBuffer = 240; // TODO(hlundin): Remove.
static const int kMaxBytesInBuffer = 113280; // TODO(hlundin): Remove.
// Creates a new NetEq object, starting at the sample rate |sample_rate_hz|.
// (Note that it will still change the sample rate depending on what payloads
// are being inserted; |sample_rate_hz| is just for startup configuration.)
static NetEq* Create(int sample_rate_hz);
virtual ~NetEq() {}
// Inserts a new packet into NetEq. The |receive_timestamp| is an indication
// of the time when the packet was received, and should be measured with
// the same tick rate as the RTP timestamp of the current payload.
// Returns 0 on success, -1 on failure.
virtual int InsertPacket(const WebRtcRTPHeader& rtp_header,
const uint8_t* payload,
int length_bytes,
uint32_t receive_timestamp) = 0;
// Instructs NetEq to deliver 10 ms of audio data. The data is written to
// |output_audio|, which can hold (at least) |max_length| elements.
// The number of channels that were written to the output is provided in
// the output variable |num_channels|, and each channel contains
// |samples_per_channel| elements. If more than one channel is written,
// the samples are interleaved.
// The speech type is written to |type|, if |type| is not NULL.
// Returns kOK on success, or kFail in case of an error.
virtual int GetAudio(size_t max_length, int16_t* output_audio,
int* samples_per_channel, int* num_channels,
NetEqOutputType* type) = 0;
// Associates |rtp_payload_type| with |codec| and stores the information in
// the codec database. Returns 0 on success, -1 on failure.
virtual int RegisterPayloadType(enum NetEqDecoder codec,
uint8_t rtp_payload_type) = 0;
// Provides an externally created decoder object |decoder| to insert in the
// decoder database. The decoder implements a decoder of type |codec| and
// associates it with |rtp_payload_type|. The decoder operates at the
// frequency |sample_rate_hz|. Returns kOK on success, kFail on failure.
virtual int RegisterExternalDecoder(AudioDecoder* decoder,
enum NetEqDecoder codec,
int sample_rate_hz,
uint8_t rtp_payload_type) = 0;
// Removes |rtp_payload_type| from the codec database. Returns 0 on success,
// -1 on failure.
virtual int RemovePayloadType(uint8_t rtp_payload_type) = 0;
// Sets the desired extra delay on top of what NetEq already applies due to
// current network situation. Used for synchronization with video. Returns
// true if successful, otherwise false.
virtual bool SetExtraDelay(int extra_delay_ms) = 0;
// Not implemented.
virtual int SetTargetDelay() = 0;
// Not implemented.
virtual int TargetDelay() = 0;
// Not implemented.
virtual int CurrentDelay() = 0;
// Enables playout of DTMF tones.
virtual int EnableDtmf() = 0;
// Sets the playout mode to |mode|.
virtual void SetPlayoutMode(NetEqPlayoutMode mode) = 0;
// Returns the current playout mode.
virtual NetEqPlayoutMode PlayoutMode() const = 0;
// Writes the current network statistics to |stats|. The statistics are reset
// after the call.
virtual int NetworkStatistics(NetEqNetworkStatistics* stats) = 0;
// Writes the last packet waiting times (in ms) to |waiting_times|. The number
// of values written is no more than 100, but may be smaller if the interface
// is polled again before 100 packets has arrived.
virtual void WaitingTimes(std::vector<int>* waiting_times) = 0;
// Writes the current RTCP statistics to |stats|. The statistics are reset
// and a new report period is started with the call.
virtual void GetRtcpStatistics(RtcpStatistics* stats) = 0;
// Same as RtcpStatistics(), but does not reset anything.
virtual void GetRtcpStatisticsNoReset(RtcpStatistics* stats) = 0;
// Enables post-decode VAD. When enabled, GetAudio() will return
// kOutputVADPassive when the signal contains no speech.
virtual void EnableVad() = 0;
// Disables post-decode VAD.
virtual void DisableVad() = 0;
// Returns the RTP timestamp for the last sample delivered by GetAudio().
virtual uint32_t PlayoutTimestamp() = 0;
// Not implemented.
virtual int SetTargetNumberOfChannels() = 0;
// Not implemented.
virtual int SetTargetSampleRate() = 0;
// Returns the error code for the last occurred error. If no error has
// occurred, 0 is returned.
virtual int LastError() = 0;
// Returns the error code last returned by a decoder (audio or comfort noise).
// When LastError() returns kDecoderErrorCode or kComfortNoiseErrorCode, check
// this method to get the decoder's error code.
virtual int LastDecoderError() = 0;
// Flushes both the packet buffer and the sync buffer.
virtual void FlushBuffers() = 0;
protected:
NetEq() {}
private:
DISALLOW_COPY_AND_ASSIGN(NetEq);
};
} // namespace webrtc
#endif // WEBRTC_MODULES_AUDIO_CODING_NETEQ4_INTERFACE_NETEQ_H_