voice_audio_multi/include/engine/community_models/parakeet_tdt/weights.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

75 lines
2.6 KiB
C++

#pragma once
#include "engine/framework/core/backend_weight_store.h"
#include "engine/framework/modules/attention/types.h"
#include "engine/framework/modules/conv_modules.h"
#include "engine/framework/modules/linear_module.h"
#include "engine/framework/modules/norm_modules.h"
#include "engine/community_models/parakeet_tdt/assets.h"
#include "engine/community_models/parakeet_tdt/predictor.h"
#include <memory>
#include <vector>
namespace engine::community_models::parakeet_tdt {
struct ParakeetSubsamplingLayerWeights {
engine::core::TensorValue depthwise_weight;
engine::core::TensorValue depthwise_bias;
engine::modules::Conv2dWeights pointwise;
};
struct ParakeetSubsamplingWeights {
engine::modules::Conv2dWeights conv_in;
std::vector<ParakeetSubsamplingLayerWeights> layers;
engine::modules::LinearWeights linear;
};
struct ParakeetFastConformerLayerWeights {
engine::modules::NormWeights norm_ff1;
engine::modules::LinearWeights ff1_linear1;
engine::modules::LinearWeights ff1_linear2;
engine::modules::NormWeights norm_attn;
engine::modules::AttentionWeights self_attn;
engine::core::TensorValue pos_weight;
engine::core::TensorValue pos_bias_u;
engine::core::TensorValue pos_bias_v;
engine::modules::NormWeights norm_conv;
engine::modules::LinearWeights conv_pw1;
engine::core::TensorValue conv_dw_weight;
engine::core::TensorValue conv_dw_bias;
engine::modules::LinearWeights conv_pw2;
engine::modules::NormWeights norm_ff2;
engine::modules::LinearWeights ff2_linear1;
engine::modules::LinearWeights ff2_linear2;
engine::modules::NormWeights norm_out;
};
struct ParakeetFastConformerWeights {
ParakeetSubsamplingWeights subsampling;
std::vector<ParakeetFastConformerLayerWeights> layers;
};
struct ParakeetTDTDecoderWeights {
engine::core::TensorValue embedding;
std::shared_ptr<const Predictor> predictor;
engine::modules::LinearWeights decoder_projector;
engine::modules::LinearWeights joint_enc; // encoder_projector: 1024→640
engine::modules::LinearWeights joint_head; // 640→8198
};
struct ParakeetWeights {
std::shared_ptr<engine::core::BackendWeightStore> store;
ParakeetFastConformerWeights encoder;
ParakeetTDTDecoderWeights decoder;
};
std::shared_ptr<const ParakeetWeights> load_parakeet_weights(
const ParakeetTDTAssets & assets,
ggml_backend_t backend,
engine::core::BackendType backend_type,
engine::assets::TensorStorageType matmul_storage_type,
engine::assets::TensorStorageType conv_storage_type,
size_t weight_context_bytes);
} // namespace engine::community_models::parakeet_tdt