voice_audio_multi/include/engine/models/ming3_tts/prompt.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

37 lines
913 B
C++

#pragma once
#include "engine/framework/tokenizers/llama_bpe.h"
#include <map>
#include <optional>
#include <string>
#include <vector>
namespace engine::models::ming3_tts {
struct MingPromptOptions {
std::string template_name = "tts";
std::string language;
std::string instruct;
std::string instruction_schema;
std::map<std::string, std::string> controls;
};
struct MingPrompt {
std::vector<int32_t> ids;
std::vector<size_t> speaker_positions;
int64_t latent_position = -1;
};
class MingPromptBuilder {
public:
explicit MingPromptBuilder(tokenizers::LlamaBpeTokenizerSpec tokenizer);
MingPrompt build(const std::string & text, const MingPromptOptions & options,
size_t speakers = 0, const std::string & reference_text = {}, size_t latent_steps = 0) const;
private:
tokenizers::LlamaBpeTokenizer tokenizer_;
};
} // namespace engine::models::ming3_tts