voice_audio_multi/include/engine/models/breeze_tts/session.h
Fabio c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
Initial import: audio.cpp with Orange Pi documentation
2026-11-04 19:05:22 +01:00

75 lines
2.7 KiB
C++

#pragma once
#include "engine/framework/model_spec/metadata.h"
#include "engine/framework/runtime/cache_slots.h"
#include "engine/framework/runtime/model.h"
#include "engine/framework/runtime/session_base.h"
#include "engine/framework/runtime/streaming_tts_session.h"
#include "engine/models/breeze_tts/assets.h"
#include "engine/models/breeze_tts/speech_decoder.h"
#include <cstddef>
#include <cstdint>
#include <chrono>
#include <memory>
#include <optional>
#include <vector>
namespace engine::models::breeze_tts {
class BreezeGeneratorRuntime;
struct BreezeGenerationRequest;
std::shared_ptr<engine::runtime::IVoiceModelLoader> make_breeze_tts_loader();
class BreezeTTSSession final
: public engine::runtime::RuntimeSessionBase
, public engine::runtime::IOfflineVoiceTaskSession
, public engine::runtime::StreamingTtsSessionBase {
public:
BreezeTTSSession(
engine::runtime::TaskSpec task,
engine::runtime::SessionOptions options,
std::shared_ptr<const BreezeTTSAssets> assets,
std::shared_ptr<const engine::model_spec::ModelContract> contract);
~BreezeTTSSession() override;
std::string family() const override;
engine::runtime::VoiceTaskKind task_kind() const override;
engine::runtime::RunMode run_mode() const override;
void prepare(const engine::runtime::SessionPreparationRequest & request) override;
engine::runtime::TaskResult run(const engine::runtime::TaskRequest & request) override;
private:
engine::runtime::TaskResult generate_stream(const engine::runtime::TaskRequest & request) override;
void reset_stream_state() override;
struct ReferenceCacheKey {
int sample_rate = 0;
int channels = 0;
uint64_t sample_count = 0;
uint64_t sample_hash = 0;
};
struct ReferenceCacheKeyEqual {
bool operator()(const ReferenceCacheKey & lhs, const ReferenceCacheKey & rhs) const noexcept;
};
struct ReferenceCacheEntry {
BreezeSpeechCodes codes;
};
BreezeSpeechCodes resolve_reference_codes(const engine::runtime::AudioBuffer & audio);
BreezeGenerationRequest build_generation_request(
const engine::runtime::TaskRequest & request,
const std::optional<BreezeSpeechCodes> & reference_codes,
size_t chunk_index) const;
engine::runtime::TaskSpec task_;
std::shared_ptr<const BreezeTTSAssets> assets_;
std::shared_ptr<const engine::model_spec::ModelContract> contract_;
std::unique_ptr<BreezeGeneratorRuntime> generator_;
engine::runtime::CacheSlots<ReferenceCacheKey, ReferenceCacheEntry, ReferenceCacheKeyEqual> reference_cache_;
std::optional<ReferenceCacheEntry> uncached_reference_;
};
} // namespace engine::models::breeze_tts