batchalign 0.7.1b13__tar.gz → 0.7.1b15__tar.gz

{batchalign-0.7.1b13/batchalign.egg-info → batchalign-0.7.1b15}/PKG-INFO

@@ -1,76 +1,16 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.1b13
+Version: 0.7.1b15
 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<2.2.0,>=2.0.1
-Requires-Dist: torchaudio<2.2.0,>=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>=1.7
-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-0.7.1b13 → batchalign-0.7.1b15}/batchalign/document.py

@@ -83,7 +83,6 @@ class TokenType(IntEnum):
     ANNOT = 4 # &~ject &~head
     PUNCT = 5 # ‡„,
     CORRECTION = 6 # test [= test]
-    VOCAL = 7 # test@o
 
 class CustomLineType(IntEnum):
     DEPENDENT = 0 # %com

{batchalign-0.7.1b13 → batchalign-0.7.1b15}/batchalign/formats/chat/lexer.py

@@ -93,8 +93,6 @@ class UtteranceLexer:
             self.__clauses.append((form, TokenType.FEAT))
         elif annotation_clean(form).strip() in CHAT_IGNORE:
             self.__clauses.append((annotation_clean(form).strip(), TokenType.ANNOT))
-        elif "@" in form:
-            self.__clauses.append((annotation_clean(form).strip(), TokenType.VOCAL))
         else:
             self.__clauses.append((annotation_clean(form).strip(), TokenType.REGULAR))
 

{batchalign-0.7.1b13 → batchalign-0.7.1b15}/batchalign/formats/chat/parser.py

@@ -90,10 +90,9 @@ def chat_parse_utterance(text, mor, gra, wor, additional):
     # seperate out main words by whether it should have phonation/morphology and add ending punct
     words = list(enumerate(tokens))
     lexed_words = [tok for tok in words if tok[1][1] in [TokenType.REGULAR,
-                                                         TokenType.PUNCT]]
+                                                            TokenType.PUNCT]]
     phonated_words = [tok for tok in words if tok[1][1] in [TokenType.REGULAR,
                                                             TokenType.RETRACE,
-                                                            TokenType.VOCAL,
                                                             TokenType.PUNCT,
                                                             TokenType.FP]]
     # create base forms

{batchalign-0.7.1b13 → batchalign-0.7.1b15}/batchalign/pipelines/dispatch.py

@@ -73,7 +73,7 @@ def dispatch_pipeline(pkg_str, lang, num_speakers=None, **arg_overrides):
             packages.append("disfluency")
         if "retracing" not in packages:
             packages.append("retracing")
-        if "utterance" not in packages and resolve("utterance", lang) == None and lang not in ["heb", "fra"]:
+        if "utterance" not in packages and resolve("utterance", lang) == None and lang not in ["heb", "fra", "ara"]:
             packages.append("utterance")
     if "fa" in packages:
         if "utr" not in packages:

{batchalign-0.7.1b13 → batchalign-0.7.1b15}/batchalign/pipelines/morphosyntax/ud.py

@@ -130,7 +130,8 @@ def handler(word, lang=None):
 
     # fix dash
     target = target.replace("-", "–")
-    if target == "“":
+
+    if "“" in target:
         target = word.text
 
     return f"{'' if not unknown else '0'}{word.upos.lower()}|{target}"
@@ -741,7 +742,6 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
             line_cut = i.strip(join_with_spaces=True)
             ending = '.'
 
-
         # clean the sentence
         line_cut = clean_sentence(line_cut)
 
@@ -818,12 +818,12 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
                 chunks_align = [(i,j) for i,j in chunks
                                 if len(j) != 0 and (j[0] not in ["<", "[", "&", "\x15", "(", ")"]) and (j[-1] not in ["]"])
                                                    and ("@" not in j)
-                                and j.strip() not in ENDING_PUNCT + MOR_PUNCT + CHAT_IGNORE + ["++"]]
+                                and j.strip() not in MOR_PUNCT + CHAT_IGNORE + ["++"]]
                 # hollow out anything we are trying to align, and leave everything else
                 chunks_backplate = [[j] 
                                     if not (len(j) != 0 and (j[0] not in ["<", "[", "&", "\x15", "(", ")"]) and (j[-1] not in ["]"])
                                     and ("@" not in j)
-                                            and j.strip() not in ENDING_PUNCT + MOR_PUNCT + CHAT_IGNORE + ["++"])
+                                            and j.strip() not in MOR_PUNCT + CHAT_IGNORE + ["++"])
                                     else
                                     []
                                     for i,j in chunks]
@@ -845,6 +845,14 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
                     elif isinstance(i, Extra) and i.extra_type == ExtraType.PAYLOAD:
                         # just put it back
                         chunks_backplate[i.payload].append(i.key)
+                    # we want to replace the morphology of forms that are not actually
+                    # supposed to be analyzed
+                    elif isinstance(i, Extra) and i.extra_type == ExtraType.REFERENCE:
+                        ut[i.payload].morphology = [Morphology(
+                            lemma = sents[0].tokens[i.payload].text,
+                            pos = "x",
+                            feats = ""
+                        )]
                 # resolve all the numbers and flatten
                 chunks_backplate = [j if isinstance(j, str) else ut[j].text
                                     for i in chunks_backplate
@@ -858,9 +866,9 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
                 retokenized_ut = retokenized_ut.replace("< ", "<")
                 retokenized_ut = retokenized_ut.replace(" :", ":")
                 retokenized_ut = retokenized_ut.replace(": <", ": <")
+                retokenized_ut = retokenized_ut.replace(" ↑", "↑")
                 retokenized_ut = re.sub(r"@ ?w ?p", "@wp", retokenized_ut)
                 retokenized_ut = retokenized_ut.replace(" @", "@")
-                retokenized_ut = re.sub(r" ↑", "↑", retokenized_ut)
                 # pray to everyone that it works---this will simply crash and ignore
                 # the utterance if it didn't work, so we are doing this as a sanity
                 # check rather than needing the parsed result

batchalign-0.7.1b15/batchalign/version

@@ -0,0 +1,3 @@
+0.7.1-beta.15
+June 02nd, 2024
+disable constiuency

{batchalign-0.7.1b13 → batchalign-0.7.1b15/batchalign.egg-info}/PKG-INFO

@@ -1,76 +1,16 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.1b13
+Version: 0.7.1b15
 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<2.2.0,>=2.0.1
-Requires-Dist: torchaudio<2.2.0,>=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>=1.7
-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-0.7.1b13/batchalign/version

@@ -1,3 +0,0 @@
-0.7.1-beta.13
-June 2nd, 2024
-more ud fixes