#pragma once #include "engine/framework/runtime/cache_slots.h" #include "engine/framework/runtime/session_base.h" #include "engine/framework/runtime/streaming_tts_session.h" #include "engine/models/fish_audio/assets.h" #include "engine/models/fish_audio/generator.h" #include #include #include #include #include namespace engine::models::fish_audio { class FishAudioSession final : public runtime::RuntimeSessionBase, public runtime::IOfflineVoiceTaskSession, public runtime::StreamingTtsSessionBase { public: FishAudioSession( runtime::TaskSpec task, runtime::SessionOptions options, std::shared_ptr assets); ~FishAudioSession() override; std::string family() const override; runtime::VoiceTaskKind task_kind() const override; runtime::RunMode run_mode() const override; void prepare(const runtime::SessionPreparationRequest & request) override; runtime::TaskResult run(const runtime::TaskRequest & request) override; private: runtime::TaskResult generate_stream(const runtime::TaskRequest & request) override; void reset_stream_state() override {} runtime::TaskResult synthesize(const runtime::TaskRequest & request, bool streaming); struct ReferenceCacheKey { std::string source_id; 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; }; struct ReferenceCacheEntry { engine::codecs::FishDacCodes codes; }; FishAudioRequest make_request(const runtime::TaskRequest & request) const; const engine::codecs::FishDacCodes & resolve_reference_codes(const FishAudioReference & reference); runtime::TaskSpec task_; std::shared_ptr assets_; std::unique_ptr generator_; std::optional defaults_; runtime::CacheSlots reference_cache_; std::optional uncached_reference_; }; } // namespace engine::models::fish_audio