voice_audio_multi/include/engine/models/hviske_asr/encoder.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

53 lines
1.7 KiB
C++

#pragma once
#include "engine/framework/core/execution_context.h"
#include "engine/models/hviske_asr/assets.h"
#include "engine/models/hviske_asr/frontend.h"
#include "engine/models/hviske_asr/weights.h"
#include <cstddef>
#include <cstdint>
#include <memory>
#include <vector>
namespace engine::models::hviske_asr {
struct HviskeEncodedAudio {
std::vector<float> values;
int64_t frames = 0;
int64_t valid_frames = 0;
int64_t hidden_size = 0;
};
class HviskeConformerEncoderRuntime {
public:
HviskeConformerEncoderRuntime(
std::shared_ptr<const HviskeASRAssets> assets,
std::shared_ptr<const HviskeWeights> weights,
engine::core::ExecutionContext & execution_context,
size_t graph_arena_bytes);
~HviskeConformerEncoderRuntime();
HviskeConformerEncoderRuntime(const HviskeConformerEncoderRuntime &) = delete;
HviskeConformerEncoderRuntime & operator=(const HviskeConformerEncoderRuntime &) = delete;
HviskeConformerEncoderRuntime(HviskeConformerEncoderRuntime &&) noexcept;
HviskeConformerEncoderRuntime & operator=(HviskeConformerEncoderRuntime &&) noexcept;
void prepare_capacity(int64_t input_frames, int64_t input_features);
HviskeEncodedAudio encode(const HviskeFrontendFeatures & features);
private:
struct Graph;
void ensure_graph(int64_t input_frames, int64_t input_features);
std::shared_ptr<const HviskeASRAssets> assets_;
std::shared_ptr<const HviskeWeights> weights_;
engine::core::ExecutionContext * execution_context_ = nullptr;
size_t graph_arena_bytes_ = 0;
std::unique_ptr<Graph> graph_;
std::vector<float> time_major_input_;
std::vector<float> output_scratch_;
};
} // namespace engine::models::hviske_asr