EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Implement context biasing with a Aho Corasick automata (#145)

Browse Source 
* Implement context graph

* Modify the interface to support context biasing

* Support context biasing in modified beam search; add python wrapper

* Support context biasing in python api example

* Minor fixes

* Fix context graph

* Minor fixes

* Fix tests

* Fix style

* Fix style

* Fix comments

* Minor fixes

* Add missing header

* Replace std::shared_ptr with std::unique_ptr for effciency

* Build graph in constructor

* Fix comments

* Minor fixes

* Fix docs
This commit is contained in:
Wei Kang
2023-06-16 14:26:36 +08:00
committed by GitHub
parent 1a1b9fd236
commit 8562711252
23 changed files with 515 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
sherpa-onnx/python/csrc/offline-recognizer.cc
View File

@@ -16,16 +16,17 @@ static void PybindOfflineRecognizerConfig(py::module *m) {
py::class_<PyClass>(*m, "OfflineRecognizerConfig")
.def(py::init<const OfflineFeatureExtractorConfig &,
const OfflineModelConfig &, const OfflineLMConfig &,
const std::string &, int32_t>(),
const std::string &, int32_t, float>(),
py::arg("feat_config"), py::arg("model_config"),
py::arg("lm_config") = OfflineLMConfig(),
py::arg("decoding_method") = "greedy_search",
py::arg("max_active_paths") = 4)
py::arg("max_active_paths") = 4, py::arg("context_score") = 1.5)
.def_readwrite("feat_config", &PyClass::feat_config)
.def_readwrite("model_config", &PyClass::model_config)
.def_readwrite("lm_config", &PyClass::lm_config)
.def_readwrite("decoding_method", &PyClass::decoding_method)
.def_readwrite("max_active_paths", &PyClass::max_active_paths)
.def_readwrite("context_score", &PyClass::context_score)
.def("__str__", &PyClass::ToString);
}
@@ -35,10 +36,18 @@ void PybindOfflineRecognizer(py::module *m) {
using PyClass = OfflineRecognizer;
py::class_<PyClass>(*m, "OfflineRecognizer")
.def(py::init<const OfflineRecognizerConfig &>(), py::arg("config"))
.def("create_stream", &PyClass::CreateStream)
.def("create_stream",
[](const PyClass &self) { return self.CreateStream(); })
.def(
"create_stream",
[](PyClass &self,
const std::vector<std::vector<int32_t>> &contexts_list) {
return self.CreateStream(contexts_list);
},
py::arg("contexts_list"))
.def("decode_stream", &PyClass::DecodeStream)
.def("decode_streams",
[](PyClass &self, std::vector<OfflineStream *> ss) {
[](const PyClass &self, std::vector<OfflineStream *> ss) {
self.DecodeStreams(ss.data(), ss.size());
});
}