Add online punctuation and casing prediction model for English language (#1224)

This commit is contained in:
jianyou
2024-08-06 17:33:38 +08:00
committed by GitHub
parent 52830cc910
commit 1414e4dc61
14 changed files with 874 additions and 0 deletions

View File

@@ -0,0 +1,58 @@
// sherpa-onnx/csrc/online-punctuation.h
//
// Copyright (c) 2024 Jian You (jianyou@cisco.com, Cisco Systems)
#ifndef SHERPA_ONNX_CSRC_ONLINE_PUNCTUATION_H_
#define SHERPA_ONNX_CSRC_ONLINE_PUNCTUATION_H_
#include <memory>
#include <string>
#include <vector>
#if __ANDROID_API__ >= 9
#include "android/asset_manager.h"
#include "android/asset_manager_jni.h"
#endif
#include "sherpa-onnx/csrc/online-punctuation-model-config.h"
#include "sherpa-onnx/csrc/parse-options.h"
namespace sherpa_onnx {
struct OnlinePunctuationConfig {
OnlinePunctuationModelConfig model;
OnlinePunctuationConfig() = default;
explicit OnlinePunctuationConfig(const OnlinePunctuationModelConfig &model)
: model(model) {}
void Register(ParseOptions *po);
bool Validate() const;
std::string ToString() const;
};
class OnlinePunctuationImpl;
class OnlinePunctuation {
public:
explicit OnlinePunctuation(const OnlinePunctuationConfig &config);
#if __ANDROID_API__ >= 9
OnlinePunctuation(AAssetManager *mgr,
const OnlinePunctuationConfig &config);
#endif
~OnlinePunctuation();
// Add punctuation and casing to the input text and return it.
std::string AddPunctuationWithCase(const std::string &text) const;
private:
std::unique_ptr<OnlinePunctuationImpl> impl_;
};
} // namespace sherpa_onnx
#endif // SHERPA_ONNX_CSRC_ONLINE_PUNCTUATION_H_