Add Python API for punctuation models. (#762)

This commit is contained in:
Fangjun Kuang
2024-04-13 13:28:17 +08:00
committed by GitHub
parent 329fe1aa8b
commit 68b8b88b5a
14 changed files with 136 additions and 6 deletions

View File

@@ -12,6 +12,7 @@ set(srcs
offline-model-config.cc
offline-nemo-enc-dec-ctc-model-config.cc
offline-paraformer-model-config.cc
offline-punctuation.cc
offline-recognizer.cc
offline-stream.cc
offline-tdnn-model-config.cc

View File

@@ -0,0 +1,49 @@
// sherpa-onnx/python/csrc/offline-punctuation.cc
//
// Copyright (c) 2024 Xiaomi Corporation
#include "sherpa-onnx/python/csrc/offline-punctuation.h"
#include "sherpa-onnx/csrc/offline-punctuation.h"
namespace sherpa_onnx {
static void PybindOfflinePunctuationModelConfig(py::module *m) {
using PyClass = OfflinePunctuationModelConfig;
py::class_<PyClass>(*m, "OfflinePunctuationModelConfig")
.def(py::init<>())
.def(py::init<const std::string &, int32_t, bool, const std::string &>(),
py::arg("ct_transformer"), py::arg("num_threads") = 1,
py::arg("debug") = false, py::arg("provider") = "cpu")
.def_readwrite("ct_transformer", &PyClass::ct_transformer)
.def_readwrite("num_threads", &PyClass::num_threads)
.def_readwrite("debug", &PyClass::debug)
.def_readwrite("provider", &PyClass::provider)
.def("validate", &PyClass::Validate)
.def("__str__", &PyClass::ToString);
}
static void PybindOfflinePunctuationConfig(py::module *m) {
PybindOfflinePunctuationModelConfig(m);
using PyClass = OfflinePunctuationConfig;
py::class_<PyClass>(*m, "OfflinePunctuationConfig")
.def(py::init<>())
.def(py::init<const OfflinePunctuationModelConfig &>(), py::arg("model"))
.def_readwrite("model", &PyClass::model)
.def("validate", &PyClass::Validate)
.def("__str__", &PyClass::ToString);
}
void PybindOfflinePunctuation(py::module *m) {
PybindOfflinePunctuationConfig(m);
using PyClass = OfflinePunctuation;
py::class_<PyClass>(*m, "OfflinePunctuation")
.def(py::init<const OfflinePunctuationConfig &>(), py::arg("config"),
py::call_guard<py::gil_scoped_release>())
.def("add_punctuation", &PyClass::AddPunctuation, py::arg("text"),
py::call_guard<py::gil_scoped_release>());
}
} // namespace sherpa_onnx

View File

@@ -0,0 +1,16 @@
// sherpa-onnx/python/csrc/offline-punctuation.h
//
// Copyright (c) 2024 Xiaomi Corporation
#ifndef SHERPA_ONNX_PYTHON_CSRC_OFFLINE_PUNCTUATION_H_
#define SHERPA_ONNX_PYTHON_CSRC_OFFLINE_PUNCTUATION_H_
#include "sherpa-onnx/python/csrc/sherpa-onnx.h"
namespace sherpa_onnx {
void PybindOfflinePunctuation(py::module *m);
}
#endif // SHERPA_ONNX_PYTHON_CSRC_OFFLINE_PUNCTUATION_H_

View File

@@ -14,6 +14,7 @@
#include "sherpa-onnx/python/csrc/offline-ctc-fst-decoder-config.h"
#include "sherpa-onnx/python/csrc/offline-lm-config.h"
#include "sherpa-onnx/python/csrc/offline-model-config.h"
#include "sherpa-onnx/python/csrc/offline-punctuation.h"
#include "sherpa-onnx/python/csrc/offline-recognizer.h"
#include "sherpa-onnx/python/csrc/offline-stream.h"
#include "sherpa-onnx/python/csrc/online-ctc-fst-decoder-config.h"
@@ -40,6 +41,7 @@ PYBIND11_MODULE(_sherpa_onnx, m) {
PybindWaveWriter(&m);
PybindAudioTagging(&m);
PybindOfflinePunctuation(&m);
PybindFeatures(&m);
PybindOnlineCtcFstDecoderConfig(&m);