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

Add inverse text normalization for non-streaming ASR (#1017)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-06-17 14:28:53 +08:00
committed by GitHub
parent dd69a1b56b
commit b0f7ed3ee3
13 changed files with 380 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
sherpa-onnx/python/csrc/offline-recognizer.cc
View File

@@ -17,13 +17,14 @@ static void PybindOfflineRecognizerConfig(py::module *m) {
.def(py::init<const FeatureExtractorConfig &, const OfflineModelConfig &,
const OfflineLMConfig &, const OfflineCtcFstDecoderConfig &,
const std::string &, int32_t, const std::string &, float,
float>(),
float, const std::string &, const std::string &>(),
py::arg("feat_config"), py::arg("model_config"),
py::arg("lm_config") = OfflineLMConfig(),
py::arg("ctc_fst_decoder_config") = OfflineCtcFstDecoderConfig(),
py::arg("decoding_method") = "greedy_search",
py::arg("max_active_paths") = 4, py::arg("hotwords_file") = "",
py::arg("hotwords_score") = 1.5, py::arg("blank_penalty") = 0.0)
py::arg("hotwords_score") = 1.5, py::arg("blank_penalty") = 0.0,
py::arg("rule_fsts") = "", py::arg("rule_fars") = "")
.def_readwrite("feat_config", &PyClass::feat_config)
.def_readwrite("model_config", &PyClass::model_config)
.def_readwrite("lm_config", &PyClass::lm_config)
@@ -33,6 +34,8 @@ static void PybindOfflineRecognizerConfig(py::module *m) {
.def_readwrite("hotwords_file", &PyClass::hotwords_file)
.def_readwrite("hotwords_score", &PyClass::hotwords_score)
.def_readwrite("blank_penalty", &PyClass::blank_penalty)
.def_readwrite("rule_fsts", &PyClass::rule_fsts)
.def_readwrite("rule_fars", &PyClass::rule_fars)
.def("__str__", &PyClass::ToString);
}

70
sherpa-onnx/python/sherpa_onnx/offline_recognizer.py
View File

@@ -54,6 +54,8 @@ class OfflineRecognizer(object):
debug: bool = False,
provider: str = "cpu",
model_type: str = "transducer",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -107,6 +109,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -143,6 +151,8 @@ class OfflineRecognizer(object):
hotwords_file=hotwords_file,
hotwords_score=hotwords_score,
blank_penalty=blank_penalty,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -159,6 +169,8 @@ class OfflineRecognizer(object):
decoding_method: str = "greedy_search",
debug: bool = False,
provider: str = "cpu",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -186,6 +198,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -206,6 +224,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -222,6 +242,8 @@ class OfflineRecognizer(object):
decoding_method: str = "greedy_search",
debug: bool = False,
provider: str = "cpu",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -251,6 +273,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -271,6 +299,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -287,6 +317,8 @@ class OfflineRecognizer(object):
decoding_method: str = "greedy_search",
debug: bool = False,
provider: str = "cpu",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -315,6 +347,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -335,6 +373,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -353,6 +393,8 @@ class OfflineRecognizer(object):
debug: bool = False,
provider: str = "cpu",
tail_paddings: int = -1,
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -389,6 +431,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -415,6 +463,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -431,6 +481,8 @@ class OfflineRecognizer(object):
decoding_method: str = "greedy_search",
debug: bool = False,
provider: str = "cpu",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -458,6 +510,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -478,6 +536,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config
@@ -494,6 +554,8 @@ class OfflineRecognizer(object):
decoding_method: str = "greedy_search",
debug: bool = False,
provider: str = "cpu",
rule_fsts: str = "",
rule_fars: str = "",
):
"""
Please refer to
@@ -522,6 +584,12 @@ class OfflineRecognizer(object):
True to show debug messages.
provider:
onnxruntime execution providers. Valid values are: cpu, cuda, coreml.
rule_fsts:
If not empty, it specifies fsts for inverse text normalization.
If there are multiple fsts, they are separated by a comma.
rule_fars:
If not empty, it specifies fst archives for inverse text normalization.
If there are multiple archives, they are separated by a comma.
"""
self = cls.__new__(cls)
model_config = OfflineModelConfig(
@@ -542,6 +610,8 @@ class OfflineRecognizer(object):
feat_config=feat_config,
model_config=model_config,
decoding_method=decoding_method,
rule_fsts=rule_fsts,
rule_fars=rule_fars,
)
self.recognizer = _Recognizer(recognizer_config)
self.config = recognizer_config