diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-19 00:47:55 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-19 00:47:55 +0000 |
commit | 26a029d407be480d791972afb5975cf62c9360a6 (patch) | |
tree | f435a8308119effd964b339f76abb83a57c29483 /third_party/libwebrtc/modules/audio_coding/include | |
parent | Initial commit. (diff) | |
download | firefox-26a029d407be480d791972afb5975cf62c9360a6.tar.xz firefox-26a029d407be480d791972afb5975cf62c9360a6.zip |
Adding upstream version 124.0.1.upstream/124.0.1
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'third_party/libwebrtc/modules/audio_coding/include')
-rw-r--r-- | third_party/libwebrtc/modules/audio_coding/include/audio_coding_module.h | 146 | ||||
-rw-r--r-- | third_party/libwebrtc/modules/audio_coding/include/audio_coding_module_typedefs.h | 137 |
2 files changed, 283 insertions, 0 deletions
diff --git a/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module.h b/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module.h new file mode 100644 index 0000000000..5c5bd167c0 --- /dev/null +++ b/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module.h @@ -0,0 +1,146 @@ +/* + * 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 MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_H_ +#define MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_H_ + +#include <memory> +#include <string> +#include <utility> +#include <vector> + +#include "absl/types/optional.h" +#include "api/audio_codecs/audio_encoder.h" +#include "api/function_view.h" +#include "modules/audio_coding/include/audio_coding_module_typedefs.h" + +namespace webrtc { + +// forward declarations +class AudioDecoder; +class AudioEncoder; +class AudioFrame; +struct RTPHeader; + +// Callback class used for sending data ready to be packetized +class AudioPacketizationCallback { + public: + virtual ~AudioPacketizationCallback() {} + + virtual int32_t SendData(AudioFrameType frame_type, + uint8_t payload_type, + uint32_t timestamp, + const uint8_t* payload_data, + size_t payload_len_bytes, + int64_t absolute_capture_timestamp_ms) { + // TODO(bugs.webrtc.org/10739): Deprecate the old SendData and make this one + // pure virtual. + return SendData(frame_type, payload_type, timestamp, payload_data, + payload_len_bytes); + } + virtual int32_t SendData(AudioFrameType frame_type, + uint8_t payload_type, + uint32_t timestamp, + const uint8_t* payload_data, + size_t payload_len_bytes) { + RTC_DCHECK_NOTREACHED() << "This method must be overridden, or not used."; + return -1; + } +}; + +class AudioCodingModule { + protected: + AudioCodingModule() {} + + public: + static std::unique_ptr<AudioCodingModule> Create(); + virtual ~AudioCodingModule() = default; + + // `modifier` is called exactly once with one argument: a pointer to the + // unique_ptr that holds the current encoder (which is null if there is no + // current encoder). For the duration of the call, `modifier` has exclusive + // access to the unique_ptr; it may call the encoder, steal the encoder and + // replace it with another encoder or with nullptr, etc. + virtual void ModifyEncoder( + rtc::FunctionView<void(std::unique_ptr<AudioEncoder>*)> modifier) = 0; + + // Utility method for simply replacing the existing encoder with a new one. + void SetEncoder(std::unique_ptr<AudioEncoder> new_encoder) { + ModifyEncoder([&](std::unique_ptr<AudioEncoder>* encoder) { + *encoder = std::move(new_encoder); + }); + } + + // int32_t RegisterTransportCallback() + // Register a transport callback which will be called to deliver + // the encoded buffers whenever Process() is called and a + // bit-stream is ready. + // + // Input: + // -transport : pointer to the callback class + // transport->SendData() is called whenever + // Process() is called and bit-stream is ready + // to deliver. + // + // Return value: + // -1 if the transport callback could not be registered + // 0 if registration is successful. + // + virtual int32_t RegisterTransportCallback( + AudioPacketizationCallback* transport) = 0; + + /////////////////////////////////////////////////////////////////////////// + // int32_t Add10MsData() + // Add 10MS of raw (PCM) audio data and encode it. If the sampling + // frequency of the audio does not match the sampling frequency of the + // current encoder ACM will resample the audio. If an encoded packet was + // produced, it will be delivered via the callback object registered using + // RegisterTransportCallback, and the return value from this function will + // be the number of bytes encoded. + // + // Input: + // -audio_frame : the input audio frame, containing raw audio + // sampling frequency etc. + // + // Return value: + // >= 0 number of bytes encoded. + // -1 some error occurred. + // + virtual int32_t Add10MsData(const AudioFrame& audio_frame) = 0; + + /////////////////////////////////////////////////////////////////////////// + // int SetPacketLossRate() + // Sets expected packet loss rate for encoding. Some encoders provide packet + // loss gnostic encoding to make stream less sensitive to packet losses, + // through e.g., FEC. No effects on codecs that do not provide such encoding. + // + // Input: + // -packet_loss_rate : expected packet loss rate (0 -- 100 inclusive). + // + // Return value + // -1 if failed to set packet loss rate, + // 0 if succeeded. + // + // This is only used in test code that rely on old ACM APIs. + // TODO(minyue): Remove it when possible. + virtual int SetPacketLossRate(int packet_loss_rate) = 0; + + /////////////////////////////////////////////////////////////////////////// + // statistics + // + + virtual ANAStats GetANAStats() const = 0; + + virtual int GetTargetBitrate() const = 0; +}; + +} // namespace webrtc + +#endif // MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_H_ diff --git a/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module_typedefs.h b/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module_typedefs.h new file mode 100644 index 0000000000..4b880fb633 --- /dev/null +++ b/third_party/libwebrtc/modules/audio_coding/include/audio_coding_module_typedefs.h @@ -0,0 +1,137 @@ +/* + * Copyright (c) 2011 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 MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_TYPEDEFS_H_ +#define MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_TYPEDEFS_H_ + +#include <map> + +namespace webrtc { + +/////////////////////////////////////////////////////////////////////////// +// enum ACMVADMode +// An enumerator for aggressiveness of VAD +// -VADNormal : least aggressive mode. +// -VADLowBitrate : more aggressive than "VADNormal" to save on +// bit-rate. +// -VADAggr : an aggressive mode. +// -VADVeryAggr : the most agressive mode. +// +enum ACMVADMode { + VADNormal = 0, + VADLowBitrate = 1, + VADAggr = 2, + VADVeryAggr = 3 +}; + +enum class AudioFrameType { + kEmptyFrame = 0, + kAudioFrameSpeech = 1, + kAudioFrameCN = 2, +}; + +/////////////////////////////////////////////////////////////////////////// +// +// Enumeration of Opus mode for intended application. +// +// kVoip : optimized for voice signals. +// kAudio : optimized for non-voice signals like music. +// +enum OpusApplicationMode { + kVoip = 0, + kAudio = 1, +}; + +// Statistics for calls to AudioCodingModule::PlayoutData10Ms(). +struct AudioDecodingCallStats { + AudioDecodingCallStats() + : calls_to_silence_generator(0), + calls_to_neteq(0), + decoded_normal(0), + decoded_neteq_plc(0), + decoded_codec_plc(0), + decoded_cng(0), + decoded_plc_cng(0), + decoded_muted_output(0) {} + + int calls_to_silence_generator; // Number of calls where silence generated, + // and NetEq was disengaged from decoding. + int calls_to_neteq; // Number of calls to NetEq. + int decoded_normal; // Number of calls where audio RTP packet decoded. + int decoded_neteq_plc; // Number of calls resulted in NetEq PLC. + int decoded_codec_plc; // Number of calls resulted in codec PLC. + int decoded_cng; // Number of calls where comfort noise generated due to DTX. + int decoded_plc_cng; // Number of calls resulted where PLC faded to CNG. + int decoded_muted_output; // Number of calls returning a muted state output. +}; + +// NETEQ statistics. +struct NetworkStatistics { + // current jitter buffer size in ms + uint16_t currentBufferSize; + // preferred (optimal) buffer size in ms + uint16_t preferredBufferSize; + // adding extra delay due to "peaky jitter" + bool jitterPeaksFound; + // Stats below correspond to similarly-named fields in the WebRTC stats spec. + // https://w3c.github.io/webrtc-stats/#dom-rtcinboundrtpstreamstats + uint64_t totalSamplesReceived; + uint64_t concealedSamples; + uint64_t silentConcealedSamples; + uint64_t concealmentEvents; + uint64_t jitterBufferDelayMs; + uint64_t jitterBufferTargetDelayMs; + uint64_t jitterBufferMinimumDelayMs; + uint64_t jitterBufferEmittedCount; + uint64_t insertedSamplesForDeceleration; + uint64_t removedSamplesForAcceleration; + uint64_t fecPacketsReceived; + uint64_t fecPacketsDiscarded; + // Stats below correspond to similarly-named fields in the WebRTC stats spec. + // https://w3c.github.io/webrtc-stats/#dom-rtcreceivedrtpstreamstats + uint64_t packetsDiscarded; + // Stats below DO NOT correspond directly to anything in the WebRTC stats + // fraction (of original stream) of synthesized audio inserted through + // expansion (in Q14) + uint16_t currentExpandRate; + // fraction (of original stream) of synthesized speech inserted through + // expansion (in Q14) + uint16_t currentSpeechExpandRate; + // fraction of synthesized speech inserted through pre-emptive expansion + // (in Q14) + uint16_t currentPreemptiveRate; + // fraction of data removed through acceleration (in Q14) + uint16_t currentAccelerateRate; + // fraction of data coming from secondary decoding (in Q14) + uint16_t currentSecondaryDecodedRate; + // Fraction of secondary data, including FEC and RED, that is discarded (in + // Q14). Discarding of secondary data can be caused by the reception of the + // primary data, obsoleting the secondary data. It can also be caused by early + // or late arrival of secondary data. + uint16_t currentSecondaryDiscardedRate; + // average packet waiting time in the jitter buffer (ms) + int meanWaitingTimeMs; + // max packet waiting time in the jitter buffer (ms) + int maxWaitingTimeMs; + // count of the number of buffer flushes + uint64_t packetBufferFlushes; + // number of samples expanded due to delayed packets + uint64_t delayedPacketOutageSamples; + // arrival delay of incoming packets + uint64_t relativePacketArrivalDelayMs; + // number of audio interruptions + int32_t interruptionCount; + // total duration of audio interruptions + int32_t totalInterruptionDurationMs; +}; + +} // namespace webrtc + +#endif // MODULES_AUDIO_CODING_INCLUDE_AUDIO_CODING_MODULE_TYPEDEFS_H_ |