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
34 lines
943 B
C++
34 lines
943 B
C++
#pragma once
|
|
|
|
#include "engine/framework/audio/nemo_mel_frontend.h"
|
|
#include "engine/framework/runtime/session.h"
|
|
#include "engine/models/nemotron_asr/assets.h"
|
|
|
|
#include <cstdint>
|
|
#include <memory>
|
|
#include <vector>
|
|
|
|
namespace engine::models::nemotron_asr {
|
|
|
|
struct NemotronFrontendFeatures {
|
|
std::vector<float> values;
|
|
int64_t frames = 0;
|
|
int64_t valid_frames = 0;
|
|
int64_t feature_dim = 0;
|
|
};
|
|
|
|
class NemotronFrontend {
|
|
public:
|
|
explicit NemotronFrontend(std::shared_ptr<const NemotronASRAssets> assets);
|
|
|
|
NemotronFrontendFeatures extract(
|
|
const engine::runtime::AudioBuffer & audio,
|
|
bool center) const;
|
|
std::vector<float> prepare_waveform(const engine::runtime::AudioBuffer & audio) const;
|
|
NemotronFrontendFeatures extract_waveform(const std::vector<float> & waveform, bool center) const;
|
|
|
|
private:
|
|
engine::audio::NemoMelFrontend frontend_;
|
|
};
|
|
|
|
} // namespace engine::models::nemotron_asr
|