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
61 lines
2.3 KiB
C++
61 lines
2.3 KiB
C++
#pragma once
|
|
|
|
#include "engine/models/vibevoice/connector.h"
|
|
#include "engine/models/vibevoice/decoder.h"
|
|
#include "engine/models/vibevoice/diffusion_head.h"
|
|
#include "engine/models/vibevoice/tokenizer_audio.h"
|
|
#include "engine/models/vibevoice/tokenizer_text.h"
|
|
#include "engine/models/vibevoice/types.h"
|
|
|
|
#include <cstdint>
|
|
#include <functional>
|
|
#include <vector>
|
|
|
|
namespace engine::models::vibevoice {
|
|
|
|
struct VibeVoicePreparedPrompt {
|
|
VibeVoicePromptEncoding encoding;
|
|
std::vector<float> embeddings;
|
|
int64_t steps = 0;
|
|
int64_t hidden_size = 0;
|
|
uint64_t next_rng_index = 0;
|
|
};
|
|
|
|
VibeVoicePreparedPrompt prepare_vibevoice_prompt(
|
|
const VibeVoiceRequest & request,
|
|
const VibeVoiceTextTokenizer & text_tokenizer,
|
|
const VibeVoiceTokenizerWeightsRuntime & audio_tokenizer,
|
|
const VibeVoiceConnectorWeightsRuntime & connector,
|
|
const VibeVoiceQwen2WeightsRuntime & decoder,
|
|
uint64_t seed,
|
|
uint64_t start_rng_index,
|
|
const std::vector<float> * prompt_noise_values = nullptr);
|
|
|
|
VibeVoiceResult generate_vibevoice(
|
|
const VibeVoiceRequest & request,
|
|
const VibeVoiceTextTokenizer & text_tokenizer,
|
|
const VibeVoiceTokenizerWeightsRuntime & audio_tokenizer,
|
|
const VibeVoiceConnectorWeightsRuntime & connector,
|
|
const VibeVoiceQwen2WeightsRuntime & decoder,
|
|
const VibeVoiceDiffusionHeadWeightsRuntime & diffusion_head);
|
|
|
|
VibeVoiceResult generate_vibevoice(
|
|
const VibeVoiceRequest & request,
|
|
const VibeVoiceTextTokenizer & text_tokenizer,
|
|
const VibeVoiceTokenizerWeightsRuntime & audio_tokenizer,
|
|
const VibeVoiceConnectorWeightsRuntime & connector,
|
|
const VibeVoiceQwen2WeightsRuntime & decoder,
|
|
const VibeVoiceDiffusionHeadWeightsRuntime & diffusion_head,
|
|
VibeVoiceQwen2CachedState & positive_cache,
|
|
VibeVoiceQwen2CachedState & negative_cache,
|
|
const std::function<void(const runtime::AudioBuffer &)> & on_audio = {});
|
|
|
|
std::vector<VibeVoiceResult> generate_vibevoice_batch(
|
|
const std::vector<VibeVoiceRequest> & requests,
|
|
const VibeVoiceTextTokenizer & text_tokenizer,
|
|
const VibeVoiceTokenizerWeightsRuntime & audio_tokenizer,
|
|
const VibeVoiceConnectorWeightsRuntime & connector,
|
|
const VibeVoiceQwen2WeightsRuntime & decoder,
|
|
const VibeVoiceDiffusionHeadWeightsRuntime & diffusion_head);
|
|
|
|
} // namespace engine::models::vibevoice
|