voice_audio_multi/include/engine/framework/conditioners/musiccoca_runtime.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

35 lines
1.2 KiB
C++

#pragma once
#include "engine/framework/core/execution_context.h"
#include "engine/framework/assets/tensor_source.h"
#include "engine/framework/io/json.h"
#include <array>
#include <filesystem>
#include <memory>
#include <string>
#include <vector>
namespace engine::conditioners {
class MusicCoCaRuntime {
public:
MusicCoCaRuntime(core::ExecutionContext & execution,
std::shared_ptr<const assets::TensorSource> weights,
io::json::Value graphs, const std::filesystem::path & tokenizer);
~MusicCoCaRuntime();
std::vector<float> embed_text(const std::string & text);
std::vector<float> map_text(const std::vector<float> & embedding, const std::vector<float> & noise);
std::vector<float> map_text(const std::vector<float> & embedding, uint32_t seed = 0);
std::vector<float> embed_features(const std::vector<float> & log_mel);
std::vector<float> preprocess_audio(const std::vector<float> & mono_16khz);
std::vector<float> embed_audio(const std::vector<float> & mono_16khz);
std::array<int32_t, 12> quantize(const std::vector<float> & embedding);
private:
class Impl;
std::unique_ptr<Impl> impl_;
};
} // namespace engine::conditioners