voice_audio_multi/include/engine/community_models/glm_tts/flow.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

54 lines
1.4 KiB
C++

#pragma once
#include "engine/community_models/glm_tts/assets.h"
#include "engine/framework/assets/tensor_source.h"
#include "engine/framework/core/backend.h"
#include <cstdint>
#include <memory>
#include <vector>
namespace engine::models::glm_tts {
struct GlmTTSFlowInput {
std::vector<int32_t> speech_tokens;
std::vector<float> prompt_mel;
std::vector<float> speaker_embedding;
std::vector<float> initial_noise;
int64_t prompt_frames = 0;
int inference_steps = 10;
float cfg_rate = 0.7F;
};
struct GlmTTSFlowOutput {
std::vector<float> mel;
int64_t frames = 0;
int64_t prompt_frames = 0;
};
class GlmTTSFlowRuntime {
public:
GlmTTSFlowRuntime() = default;
GlmTTSFlowRuntime(
std::shared_ptr<const assets::TensorSource> source,
core::BackendConfig backend,
assets::TensorStorageType storage_type,
GlmTTSFlowConfig config);
~GlmTTSFlowRuntime();
GlmTTSFlowRuntime(GlmTTSFlowRuntime &&) noexcept;
GlmTTSFlowRuntime & operator=(GlmTTSFlowRuntime &&) noexcept;
GlmTTSFlowRuntime(const GlmTTSFlowRuntime &) = delete;
GlmTTSFlowRuntime & operator=(const GlmTTSFlowRuntime &) = delete;
GlmTTSFlowOutput generate(const GlmTTSFlowInput & input) const;
void release_graph();
private:
struct State;
GlmTTSFlowConfig config_;
std::shared_ptr<State> state_;
};
} // namespace engine::models::glm_tts