Files
voice-cat/core/src/codec/opus_codec.h

128 lines
4.4 KiB
C
Raw Normal View History

/*
* codec/opus_codec.h Opus encode/decode (libopus 1.6).
*
* Design: docs/voice.md §34. Per-channel AudioConfig (mono/stereo, bitrate, frame size,
* FEC, DTX, complexity). The server relays Opus payloads unmodified (no transcode).
*/
#ifndef VOICECAT_CODEC_OPUS_CODEC_H
#define VOICECAT_CODEC_OPUS_CODEC_H
#include <cstdint>
#include <vector>
#ifdef VOICECAT_HAS_OPUS
#include <opus/opus.h>
#endif
namespace voicecat::codec {
feat(M3): multi-stream & per-channel tuning Implements docs/roadmap.md M3: multiple concurrent streams per user (MIC + SCREEN_AUDIO + AUX_DEVICE), independent per-stream receiver gain/mute/noise- reduction, talk indicators, and enforced per-channel Opus configurability (mono/stereo, bitrate, frame size, FEC/DTX, application). Bugs fixed along the way (found while implementing, not pre-existing scope): - Server hard-coded stream_id=1 for every announce, so a second stream from the same user silently overwrote the first in SessionRegistry::set_user_stream. Now a per-session counter (ConnSession::next_stream_id_); handle_stream_stop validates against announced_stream_ids_ before clearing. - Client dropped mode/dtx/complexity/application from effective_audio even for the single M2 stream -- only sample_rate/bitrate_bps/frame_ms/fec were ever applied to OpusParams. Fixed on both the send (handle_stream_announce_result) and receive (sync_remote_streams) paths via a shared opus_params_from_audio_config() helper. - OpusEncoder always used OPUS_APPLICATION_VOIP; added OpusParams::application and wired it through. - on_playback's per-stream decode passed the wrong frame_size to opus_decode (total samples instead of samples-per-channel), which would have overflowed the decode buffer for any stereo stream. - teardown_voice() raced when called concurrently from run_io()'s own cleanup and from disconnect() on a different thread -- both could see udp_thread_/talk_timer_thread_ as joinable() at once and race to join() the same std::thread (intermittent std::system_error under ctest). Fixed with a teardown_mu_ guard instead of carrying the flake forward. New: - Per-channel AudioConfig: SessionRegistry now seeds Lobby (mono/24kbps/VOIP/ FEC+DTX) and a new "Music Room" channel (stereo/128kbps/AUDIO/no DTX); handle_stream_announce enforces the channel's config, clamping (not overriding) bitrate_bps to its ceiling. - core/src/core/client.h/.cpp: local-stream state is now a std::unordered_map<int, LocalStream> keyed by vc_stream_kind, with request_id-correlated announce/result handling (request_id already round-tripped on the wire; just wasn't read before). on_capture_frame is kind-aware and upmixes mono capture to stereo when a stream's config calls for it. set_self_mute's mic_muted now only gates the MIC kind. NS is wired through set_remote_stream. New run_talk_timer() thread emits VC_EVENT_TALK_STATE from both remote and local edge detection. - core/src/audio/audio_engine.h/.cpp: kind-keyed injection taps (inject_capture), stereo-to-mono downmix at the decode/mix boundary, RemoteStream gains recv_ns (lazy ApmProcessor) + noise_reduction_enabled and last_voice_ms/talking; new set_stream_noise_reduction() and poll_talk_transitions(). - core/src/session/session.h/.cpp: Stream now carries the full AudioConfig, not just sample_rate/frame_ms. - New additive C ABI (core/include/voicecat.h): vc_audio_config + vc_get_stream_audio_config (effective Opus config for any stream you own or a peer's); vc_test_inject_capture (test-only synthetic PCM injection, clearly marked, mirrors AudioEngine::inject_capture). - tests/test_m3_multistream.cpp: the M3 exit criterion through the real ABI (mirrors test_voice_client_abi.cpp's approach, not raw sockets) -- two concurrent local streams, independent gain/mute/NS control, per-channel config divergence via vc_get_stream_audio_config, talk indicators. Explicitly out of scope for this pass (tracked in PROGRESS.md, not silently dropped): VAD/PTT input gate + device enumeration; real WASAPI loopback capture for SCREEN_AUDIO (synthetic injection only); true stereo playback output (AudioEngine's mixer/output device stays mono -- Opus itself is fully stereo-correct on the wire). ctest --test-dir build/m1-dev: 11/11 green, verified across 3 consecutive full-suite runs plus 8 standalone runs of the new test. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-16 14:12:37 +02:00
// Mirrors voicecat::v1::OpusApplication (core/proto/voicecat.proto) without depending on
// generated protobuf headers from this low-level codec module.
enum class OpusApplication {
Voip = 0, // speech, optimized for low-rate intelligibility
Audio = 1, // music/screen-audio, optimized for fidelity
LowDelay = 2, // monitoring, minimal algorithmic delay
};
struct OpusParams {
feat(M3): multi-stream & per-channel tuning Implements docs/roadmap.md M3: multiple concurrent streams per user (MIC + SCREEN_AUDIO + AUX_DEVICE), independent per-stream receiver gain/mute/noise- reduction, talk indicators, and enforced per-channel Opus configurability (mono/stereo, bitrate, frame size, FEC/DTX, application). Bugs fixed along the way (found while implementing, not pre-existing scope): - Server hard-coded stream_id=1 for every announce, so a second stream from the same user silently overwrote the first in SessionRegistry::set_user_stream. Now a per-session counter (ConnSession::next_stream_id_); handle_stream_stop validates against announced_stream_ids_ before clearing. - Client dropped mode/dtx/complexity/application from effective_audio even for the single M2 stream -- only sample_rate/bitrate_bps/frame_ms/fec were ever applied to OpusParams. Fixed on both the send (handle_stream_announce_result) and receive (sync_remote_streams) paths via a shared opus_params_from_audio_config() helper. - OpusEncoder always used OPUS_APPLICATION_VOIP; added OpusParams::application and wired it through. - on_playback's per-stream decode passed the wrong frame_size to opus_decode (total samples instead of samples-per-channel), which would have overflowed the decode buffer for any stereo stream. - teardown_voice() raced when called concurrently from run_io()'s own cleanup and from disconnect() on a different thread -- both could see udp_thread_/talk_timer_thread_ as joinable() at once and race to join() the same std::thread (intermittent std::system_error under ctest). Fixed with a teardown_mu_ guard instead of carrying the flake forward. New: - Per-channel AudioConfig: SessionRegistry now seeds Lobby (mono/24kbps/VOIP/ FEC+DTX) and a new "Music Room" channel (stereo/128kbps/AUDIO/no DTX); handle_stream_announce enforces the channel's config, clamping (not overriding) bitrate_bps to its ceiling. - core/src/core/client.h/.cpp: local-stream state is now a std::unordered_map<int, LocalStream> keyed by vc_stream_kind, with request_id-correlated announce/result handling (request_id already round-tripped on the wire; just wasn't read before). on_capture_frame is kind-aware and upmixes mono capture to stereo when a stream's config calls for it. set_self_mute's mic_muted now only gates the MIC kind. NS is wired through set_remote_stream. New run_talk_timer() thread emits VC_EVENT_TALK_STATE from both remote and local edge detection. - core/src/audio/audio_engine.h/.cpp: kind-keyed injection taps (inject_capture), stereo-to-mono downmix at the decode/mix boundary, RemoteStream gains recv_ns (lazy ApmProcessor) + noise_reduction_enabled and last_voice_ms/talking; new set_stream_noise_reduction() and poll_talk_transitions(). - core/src/session/session.h/.cpp: Stream now carries the full AudioConfig, not just sample_rate/frame_ms. - New additive C ABI (core/include/voicecat.h): vc_audio_config + vc_get_stream_audio_config (effective Opus config for any stream you own or a peer's); vc_test_inject_capture (test-only synthetic PCM injection, clearly marked, mirrors AudioEngine::inject_capture). - tests/test_m3_multistream.cpp: the M3 exit criterion through the real ABI (mirrors test_voice_client_abi.cpp's approach, not raw sockets) -- two concurrent local streams, independent gain/mute/NS control, per-channel config divergence via vc_get_stream_audio_config, talk indicators. Explicitly out of scope for this pass (tracked in PROGRESS.md, not silently dropped): VAD/PTT input gate + device enumeration; real WASAPI loopback capture for SCREEN_AUDIO (synthetic injection only); true stereo playback output (AudioEngine's mixer/output device stays mono -- Opus itself is fully stereo-correct on the wire). ctest --test-dir build/m1-dev: 11/11 green, verified across 3 consecutive full-suite runs plus 8 standalone runs of the new test. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-16 14:12:37 +02:00
uint32_t sample_rate = 48000;
uint32_t bitrate_bps = 24000;
uint32_t frame_ms = 20;
bool stereo = false;
bool fec = true;
bool dtx = false;
uint32_t complexity = 10;
uint32_t expected_packet_loss = 0; // % 0..100
OpusApplication application = OpusApplication::Voip;
bool dred = false;
};
// Returns frame_samples for a given sample_rate + frame_ms.
inline int opus_frame_samples(const OpusParams& p) {
return static_cast<int>(p.sample_rate / 1000 * p.frame_ms);
}
class OpusEncoder {
public:
OpusEncoder() = default;
~OpusEncoder() { destroy(); }
OpusEncoder(const OpusEncoder&) = delete;
OpusEncoder& operator=(const OpusEncoder&) = delete;
// Initialise with the given params. Must be called before encode().
// Returns true on success; check error_string() on failure.
bool init(const OpusParams& p);
// Encode one frame of PCM (frame_ms ms @ sample_rate Hz, mono or stereo).
// pcm: interleaved int16 samples (frame_samples * channels samples).
// out_buf: caller-allocated output buffer (recommend >= 4000 bytes).
// Returns number of bytes written to out_buf, or -1 on error.
int encode(const int16_t* pcm, int frame_samples, uint8_t* out_buf, int out_cap);
void destroy();
bool valid() const { return enc_ != nullptr; }
int frame_samples()const { return frame_samples_; }
int channels() const { return channels_; }
const char* error_string() const { return err_; }
private:
#ifdef VOICECAT_HAS_OPUS
::OpusEncoder* enc_ = nullptr;
#else
void* enc_ = nullptr;
#endif
int frame_samples_ = 0;
int channels_ = 1;
const char* err_ = nullptr;
};
class OpusDecoder {
public:
OpusDecoder() = default;
~OpusDecoder() { destroy(); }
OpusDecoder(const OpusDecoder&) = delete;
OpusDecoder& operator=(const OpusDecoder&) = delete;
// Initialise. Must be called before decode().
bool init(const OpusParams& p);
// Decode one Opus packet into out_pcm (frame_samples * channels int16 samples).
// opus_data=nullptr, len=0 → PLC (free, always enabled by libopus).
// fec=true, next valid packet in opus_data → FEC recovery from previous loss.
// Returns number of samples decoded (= frame_samples), or -1 on error.
int decode(const uint8_t* opus_data, int len, int16_t* out_pcm, int max_samples,
bool fec = false);
// Decode a lost frame using pre-parsed DRED state from the next received packet.
// dred_offset=0 means the frame immediately before the next packet.
// Returns frame_samples on success, -1 if DRED unavailable or decode failed.
#ifdef VOICECAT_HAS_OPUS
int decode_dred(::OpusDRED* dred, int32_t dred_offset, int16_t* out_pcm, int max_samples);
::OpusDecoder* raw() const { return dec_; }
#endif
void destroy();
bool valid() const { return dec_ != nullptr; }
int frame_samples()const { return frame_samples_; }
int channels() const { return channels_; }
const char* error_string() const { return err_; }
private:
#ifdef VOICECAT_HAS_OPUS
::OpusDecoder* dec_ = nullptr;
#else
void* dec_ = nullptr;
#endif
int frame_samples_ = 0;
int channels_ = 1;
const char* err_ = nullptr;
};
} // namespace voicecat::codec
#endif // VOICECAT_CODEC_OPUS_CODEC_H