batchalign 0.7.6a2__tar.gz → 0.7.6a3__tar.gz

{batchalign-0.7.6a2/batchalign.egg-info → batchalign-0.7.6a3}/PKG-INFO

@@ -1,76 +1,16 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.6a2
+Version: 0.7.6a3
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
 Classifier: Development Status :: 3 - Alpha
 Classifier: Topic :: Utilities
 Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: pydantic>=2.4
-Requires-Dist: nltk>=3.8
-Requires-Dist: praatio<6.1.0,>=6.0.0
-Requires-Dist: torch>=2.0.1
-Requires-Dist: torchaudio>=2.1.0
-Requires-Dist: pyAudioAnalysis==0.3.14
-Requires-Dist: hmmlearn==0.3.0
-Requires-Dist: eyed3~=0.9.7
-Requires-Dist: pydub<0.26.0,>=0.25.1
-Requires-Dist: imblearn
-Requires-Dist: plotly>=5.18.0
-Requires-Dist: transformers>=4.37
-Requires-Dist: tokenizers>=0.14.1
-Requires-Dist: pycountry>=22.3
-Requires-Dist: stanza[transformers]>=1.9.1
-Requires-Dist: scipy~=1.11
-Requires-Dist: rev_ai>=2.18.0
-Requires-Dist: rich~=13.6
-Requires-Dist: click~=8.1
-Requires-Dist: matplotlib<4.0.0,>=3.8.0
-Requires-Dist: pyfiglet==1.0.2
-Requires-Dist: soundfile~=0.12.0
-Requires-Dist: rich-click>=1.7.0
-Requires-Dist: typing-extensions
 Provides-Extra: dev
-Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train
-Requires-Dist: accelerate~=0.27; extra == "train"
 Provides-Extra: speaker
-Requires-Dist: nemo-toolkit~=1.21.0; extra == "speaker"
-Requires-Dist: omegaconf~=2.3.0; extra == "speaker"
-Requires-Dist: pydub~=0.25.0; extra == "speaker"
-Requires-Dist: braceexpand; extra == "speaker"
-Requires-Dist: editdistance; extra == "speaker"
-Requires-Dist: g2p_en; extra == "speaker"
-Requires-Dist: ipywidgets; extra == "speaker"
-Requires-Dist: jiwer; extra == "speaker"
-Requires-Dist: kaldi-python-io; extra == "speaker"
-Requires-Dist: kaldiio; extra == "speaker"
-Requires-Dist: lhotse>=1.20.0; extra == "speaker"
-Requires-Dist: librosa>=0.10.0; extra == "speaker"
-Requires-Dist: marshmallow; extra == "speaker"
-Requires-Dist: matplotlib; extra == "speaker"
-Requires-Dist: packaging; extra == "speaker"
-Requires-Dist: pyannote.core; extra == "speaker"
-Requires-Dist: pyannote.metrics; extra == "speaker"
-Requires-Dist: pydub; extra == "speaker"
-Requires-Dist: pyloudnorm; extra == "speaker"
-Requires-Dist: resampy; extra == "speaker"
-Requires-Dist: ruamel.yaml; extra == "speaker"
-Requires-Dist: scipy>=0.14; extra == "speaker"
-Requires-Dist: soundfile; extra == "speaker"
-Requires-Dist: sox; extra == "speaker"
-Requires-Dist: texterrors; extra == "speaker"
-Requires-Dist: hydra-core<=1.3.2,>1.3; extra == "speaker"
-Requires-Dist: omegaconf<=2.3; extra == "speaker"
-Requires-Dist: pytorch-lightning>=2.2.1; extra == "speaker"
-Requires-Dist: torchmetrics>=0.11.0; extra == "speaker"
-Requires-Dist: transformers>=4.36.0; extra == "speaker"
-Requires-Dist: wandb; extra == "speaker"
-Requires-Dist: webdataset>=0.2.86; extra == "speaker"
-Requires-Dist: sentencepiece; extra == "speaker"
-Requires-Dist: youtokentome; extra == "speaker"
+License-File: LICENSE
 
 # TalkBank | Batchalign2
 

{batchalign-0.7.6a2 → batchalign-0.7.6a3}/batchalign/models/whisper/infer_asr.py

@@ -191,6 +191,8 @@ class WhisperASRModel(object):
 
             if element["type"] == "text":
                 pl = element["payload"].strip()
+                pl = pl.replace("「", "")
+                pl = pl.replace("」", "")
                 before = re.findall(r"^\W+", pl)
                 after = re.findall(r"\W+$", pl)
                 texts = []

batchalign-0.7.6a3/batchalign/version

@@ -0,0 +1,3 @@
+0.7.6-alpha.3
+September 28th, 2024
+erase quote marks for cantonese

{batchalign-0.7.6a2 → batchalign-0.7.6a3/batchalign.egg-info}/PKG-INFO

@@ -1,76 +1,16 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.6a2
+Version: 0.7.6a3
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
 Classifier: Development Status :: 3 - Alpha
 Classifier: Topic :: Utilities
 Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: pydantic>=2.4
-Requires-Dist: nltk>=3.8
-Requires-Dist: praatio<6.1.0,>=6.0.0
-Requires-Dist: torch>=2.0.1
-Requires-Dist: torchaudio>=2.1.0
-Requires-Dist: pyAudioAnalysis==0.3.14
-Requires-Dist: hmmlearn==0.3.0
-Requires-Dist: eyed3~=0.9.7
-Requires-Dist: pydub<0.26.0,>=0.25.1
-Requires-Dist: imblearn
-Requires-Dist: plotly>=5.18.0
-Requires-Dist: transformers>=4.37
-Requires-Dist: tokenizers>=0.14.1
-Requires-Dist: pycountry>=22.3
-Requires-Dist: stanza[transformers]>=1.9.1
-Requires-Dist: scipy~=1.11
-Requires-Dist: rev_ai>=2.18.0
-Requires-Dist: rich~=13.6
-Requires-Dist: click~=8.1
-Requires-Dist: matplotlib<4.0.0,>=3.8.0
-Requires-Dist: pyfiglet==1.0.2
-Requires-Dist: soundfile~=0.12.0
-Requires-Dist: rich-click>=1.7.0
-Requires-Dist: typing-extensions
 Provides-Extra: dev
-Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train
-Requires-Dist: accelerate~=0.27; extra == "train"
 Provides-Extra: speaker
-Requires-Dist: nemo-toolkit~=1.21.0; extra == "speaker"
-Requires-Dist: omegaconf~=2.3.0; extra == "speaker"
-Requires-Dist: pydub~=0.25.0; extra == "speaker"
-Requires-Dist: braceexpand; extra == "speaker"
-Requires-Dist: editdistance; extra == "speaker"
-Requires-Dist: g2p_en; extra == "speaker"
-Requires-Dist: ipywidgets; extra == "speaker"
-Requires-Dist: jiwer; extra == "speaker"
-Requires-Dist: kaldi-python-io; extra == "speaker"
-Requires-Dist: kaldiio; extra == "speaker"
-Requires-Dist: lhotse>=1.20.0; extra == "speaker"
-Requires-Dist: librosa>=0.10.0; extra == "speaker"
-Requires-Dist: marshmallow; extra == "speaker"
-Requires-Dist: matplotlib; extra == "speaker"
-Requires-Dist: packaging; extra == "speaker"
-Requires-Dist: pyannote.core; extra == "speaker"
-Requires-Dist: pyannote.metrics; extra == "speaker"
-Requires-Dist: pydub; extra == "speaker"
-Requires-Dist: pyloudnorm; extra == "speaker"
-Requires-Dist: resampy; extra == "speaker"
-Requires-Dist: ruamel.yaml; extra == "speaker"
-Requires-Dist: scipy>=0.14; extra == "speaker"
-Requires-Dist: soundfile; extra == "speaker"
-Requires-Dist: sox; extra == "speaker"
-Requires-Dist: texterrors; extra == "speaker"
-Requires-Dist: hydra-core<=1.3.2,>1.3; extra == "speaker"
-Requires-Dist: omegaconf<=2.3; extra == "speaker"
-Requires-Dist: pytorch-lightning>=2.2.1; extra == "speaker"
-Requires-Dist: torchmetrics>=0.11.0; extra == "speaker"
-Requires-Dist: transformers>=4.36.0; extra == "speaker"
-Requires-Dist: wandb; extra == "speaker"
-Requires-Dist: webdataset>=0.2.86; extra == "speaker"
-Requires-Dist: sentencepiece; extra == "speaker"
-Requires-Dist: youtokentome; extra == "speaker"
+License-File: LICENSE
 
 # TalkBank | Batchalign2
 

{batchalign-0.7.6a2 → batchalign-0.7.6a3}/batchalign.egg-info/SOURCES.txt

@@ -70,7 +70,6 @@ batchalign/pipelines/cleanup/support/test.test
 batchalign/pipelines/fa/__init__.py
 batchalign/pipelines/fa/whisper_fa.py
 batchalign/pipelines/morphosyntax/__init__.py
-batchalign/pipelines/morphosyntax/coref.py
 batchalign/pipelines/morphosyntax/ud.py
 batchalign/pipelines/morphosyntax/fr/case.py
 batchalign/pipelines/morphosyntax/ja/verbforms.py

batchalign-0.7.6a2/batchalign/pipelines/morphosyntax/coref.py

@@ -1,45 +0,0 @@
-import stanza
-from batchalign.utils.dp import PayloadTarget, ReferenceTarget, Match, align
-from warnings import warn
-from batchalign.document import *
-from batchalign.constants import *
-from batchalign.pipelines.base import *
-from batchalign.formats.chat.parser import chat_parse_utterance
-
-from batchalign.utils.dp import *
-
-
-
-class CorefEngine(BatchalignEngine):
-    tasks = [ Task.COREF ]
-
-    def process(self, doc, **kwargs):
-        if "eng" not in doc.langs:
-            warn("Coreference resolution is only supported for English documents.")
-            return 
-
-        detokenized = " ".join([i.strip(include_retrace=True, include_fp=True) for i in doc.content if isinstance(i, Utterance)])
-        pipeline = stanza.Pipeline(lang="en", processors="tokenize, coref")
-
-        coref_chains = pipeline(detokenized).sentences
-        coref_chains = [(j.text,
-                        [Coref(start=chain.is_start,
-                                end=chain.is_end,
-                                chain=chain.chain.index) for chain in j.coref_chains])
-                        for i in coref_chains
-                        for j in i.words]
-
-        payloads = [PayloadTarget(i[0], i[1]) for i in coref_chains]
-        references = [ReferenceTarget(j.text, (ut_id, form_id)) for ut_id, i in enumerate(doc.content)
-                        if isinstance(i, Utterance)
-                        for form_id, j in enumerate(i.content)]
-        alignment = align(payloads, references, tqdm=False)
-
-        for i in alignment:
-            if isinstance(i, Match):
-                (ut, form) = i.reference_payload
-                doc.content[ut].content[form].coreference = i.payload
-
-        return doc
-
-

batchalign-0.7.6a2/batchalign/version

@@ -1,3 +0,0 @@
-0.7.6-alpha.2
-September 27th, 2024
-skip extra tokenizer output