batchalign 0.7.10.post2__tar.gz → 0.7.10.post5__tar.gz

{batchalign-0.7.10.post2/batchalign.egg-info → batchalign-0.7.10.post5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.10.post2
+Version: 0.7.10.post5
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
@@ -32,6 +32,7 @@ Requires-Dist: setuptools
 Requires-Dist: soundfile~=0.12.0
 Requires-Dist: rich-click>=1.7.0
 Requires-Dist: typing-extensions
+Requires-Dist: num2words
 Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train

{batchalign-0.7.10.post2 → batchalign-0.7.10.post5}/batchalign/pipelines/asr/utils.py

@@ -4,6 +4,10 @@ from batchalign.utils import *
 
 from batchalign.constants import ENDING_PUNCT
 
+from num2words import num2words
+import pycountry
+
+
 def retokenize(intermediate_output):
     """Retokenize the output of the ASR system from one giant blob to utterances
 
@@ -153,6 +157,17 @@ def process_generation(output, lang="eng", utterance_engine=None):
             for part in word_parts:
                 final_words.append([part.strip(), [cur, cur+div]])
                 cur += div
+
+        lang_2 = pycountry.languages.get(alpha_3=lang).alpha_2
+        def catched_num2words(i):
+            if not i.isdigit():
+                return i
+            try:
+                return num2words(i, lang=lang_2)
+            except NotImplementedError:
+                return i
+        final_words = [[catched_num2words(i), j] for i,j in final_words]
+
         # if the final words is > 300, split into n parts
         if len(final_words) > 300:
             # for each group, append

{batchalign-0.7.10.post2 → batchalign-0.7.10.post5}/batchalign/pipelines/morphosyntax/ud.py

@@ -990,7 +990,8 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
                     content.dependency = form.dependency
 
         except Exception as e:
-            warnings.warn(f"Utterance failed parsing, skipping ud tagging... line='{line}', error='{e}'.\n")
+            pass
+            # warnings.warn(f"Utterance failed parsing, skipping ud tagging... line='{line}', error='{e}'.\n")
 
     L.debug("Stanza done.")
     return doc

batchalign-0.7.10.post5/batchalign/version

@@ -0,0 +1,3 @@
+0.7.10-post.5
+Janurary 29th, 2025
+skipping warnings

{batchalign-0.7.10.post2 → batchalign-0.7.10.post5/batchalign.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: batchalign
-Version: 0.7.10.post2
+Version: 0.7.10.post5
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
@@ -32,6 +32,7 @@ Requires-Dist: setuptools
 Requires-Dist: soundfile~=0.12.0
 Requires-Dist: rich-click>=1.7.0
 Requires-Dist: typing-extensions
+Requires-Dist: num2words
 Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train

{batchalign-0.7.10.post2 → batchalign-0.7.10.post5}/batchalign.egg-info/requires.txt

@@ -22,6 +22,7 @@ setuptools
 soundfile~=0.12.0
 rich-click>=1.7.0
 typing-extensions
+num2words
 
 [dev]
 pytest

{batchalign-0.7.10.post2 → batchalign-0.7.10.post5}/setup.py

@@ -51,6 +51,7 @@ setup(
         "soundfile~=0.12.0",
         "rich-click>=1.7.0",
         "typing-extensions",
+        "num2words",
     ],
     extras_require={
         'dev': [

batchalign-0.7.10.post2/batchalign/version

@@ -1,3 +0,0 @@
-0.7.10-post.2
-Janurary 23th, 2025
-output Benchmark