voice_audio_multi/include/engine/models/fish_audio/generator.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

44 lines
1.4 KiB
C++

#pragma once
#include "engine/framework/codecs/fish_dac_codec_runtime.h"
#include "engine/models/fish_audio/ar.h"
#include "engine/models/fish_audio/prompt_builder.h"
#include "engine/models/fish_audio/tokenizer_text.h"
#include <memory>
#include <functional>
#include <optional>
namespace engine::models::fish_audio {
struct FishAudioGenerationResult {
runtime::AudioBuffer audio;
engine::codecs::FishDacCodes codes;
};
class FishAudioGenerator {
public:
FishAudioGenerator(
std::shared_ptr<const FishAudioAssets> assets,
std::unique_ptr<FishAudioDualARRuntime> ar,
std::unique_ptr<engine::codecs::FishDacCodecRuntime> codec);
~FishAudioGenerator();
engine::codecs::FishDacCodes encode_reference(const runtime::AudioBuffer & audio);
FishAudioGenerationResult generate(
const FishAudioRequest & request,
const std::vector<engine::codecs::FishDacCodes> & reference_codes,
const std::optional<FishAudioConversationTurn> & previous_turn,
bool mem_saver,
bool streaming = false,
const std::function<void(const runtime::AudioBuffer &)> & on_audio = {});
private:
std::shared_ptr<const FishAudioAssets> assets_;
FishAudioTextTokenizer tokenizer_;
FishAudioPromptBuilder prompt_builder_;
std::unique_ptr<FishAudioDualARRuntime> ar_;
std::unique_ptr<engine::codecs::FishDacCodecRuntime> codec_;
};
} // namespace engine::models::fish_audio