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
38 lines
1.2 KiB
C++
38 lines
1.2 KiB
C++
#pragma once
|
|
|
|
#include "engine/models/vibevoice_asr/assets.h"
|
|
#include "engine/models/vibevoice_asr/connector.h"
|
|
#include "engine/models/vibevoice_asr/speech_tokenizer.h"
|
|
#include "engine/models/vibevoice_asr/types.h"
|
|
|
|
#include <cstddef>
|
|
#include <memory>
|
|
#include <optional>
|
|
|
|
namespace engine::models::vibevoice_asr {
|
|
|
|
class VibeVoiceASRSpeechEncoder {
|
|
public:
|
|
VibeVoiceASRSpeechEncoder(
|
|
std::shared_ptr<const VibeVoiceASRAssets> assets,
|
|
core::BackendType backend_type,
|
|
int device,
|
|
int threads,
|
|
size_t tokenizer_weight_context_bytes,
|
|
size_t connector_weight_context_bytes,
|
|
assets::TensorStorageType tokenizer_weight_storage_type,
|
|
assets::TensorStorageType connector_weight_storage_type);
|
|
|
|
VibeVoiceASRSpeechFeatures encode(
|
|
const runtime::AudioBuffer & audio,
|
|
uint64_t seed,
|
|
uint64_t rng_offset = 0) const;
|
|
|
|
private:
|
|
std::shared_ptr<const VibeVoiceASRAssets> assets_;
|
|
std::optional<sampling::TorchCudaSamplingPolicy> sampling_policy_;
|
|
VibeVoiceTokenizerWeightsRuntime tokenizer_;
|
|
VibeVoiceConnectorWeightsRuntime connector_;
|
|
};
|
|
|
|
} // namespace engine::models::vibevoice_asr
|