batchalign 0.7.19__tar.gz → 0.7.19.post1__tar.gz

{batchalign-0.7.19/batchalign.egg-info → batchalign-0.7.19.post1}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.2
 Name: batchalign
-Version: 0.7.19
+Version: 0.7.19.post1
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
@@ -39,12 +39,14 @@ Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train
 Requires-Dist: accelerate~=0.27; extra == "train"
+Provides-Extra: docs
+Requires-Dist: mkdocs-material; extra == "docs"
+Requires-Dist: mkdocs-click; extra == "docs"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
 Dynamic: description
 Dynamic: description-content-type
-Dynamic: license-file
 Dynamic: provides-extra
 Dynamic: requires-dist
 Dynamic: summary

{batchalign-0.7.19 → batchalign-0.7.19.post1}/batchalign/cli/cli.py

@@ -79,6 +79,7 @@ def handle_verbosity(verbosity):
 
 @click.group()
 @click.pass_context
+@click.version_option(VERSION_NUMBER)
 @click.option("-v", "--verbose", type=int, count=True, default=0, help="How loquacious Batchalign should be.")
 def batchalign(ctx, verbose):
     """process .cha and/or audio files in IN_DIR and dumps them to OUT_DIR using recipe COMMAND"""

{batchalign-0.7.19 → batchalign-0.7.19.post1}/batchalign/pipelines/analysis/eval.py

@@ -14,6 +14,34 @@ from batchalign.utils.dp import align, ExtraType, Extra, Match
 import logging
 L = logging.getLogger("batchalign")
 
+def conform(x):
+    result = []
+    for i in x:
+        if "'s" in i.strip():
+            result.append(i.split("'")[0])
+            result.append("is")
+        elif "americanstyle" == i.strip():
+            result.append("american")
+            result.append("style")
+        elif "postwar" == i.strip():
+            result.append("post")
+            result.append("war")
+        elif "farmhouse" == i.strip():
+            result.append("farm")
+            result.append("house")
+        elif "aa" == i.strip():
+            result.append("a")
+            result.append("a")
+        elif "hmm" == i.strip():
+            result.append("hm")
+        elif "_" in i.strip():
+            for j in i.strip().split("_"):
+                result.append(j)
+        else:
+            result.append(i)
+
+    return result
+
 class EvaluationEngine(BatchalignEngine):
     tasks = [ Task.WER ]
 
@@ -64,6 +92,9 @@ class EvaluationEngine(BatchalignEngine):
             gold_final = gold_forms
             forms_final = forms_finished
 
+        gold_final = conform(gold_final)
+        forms_final = conform(forms_final)
+
         # dp!
         alignment = align(forms_final, gold_final, False)
 

batchalign-0.7.19.post1/batchalign/version

@@ -0,0 +1,3 @@
+0.7.19-post.1
+May 11th, 2025
+Fixes for benchmarking

{batchalign-0.7.19 → batchalign-0.7.19.post1/batchalign.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.2
 Name: batchalign
-Version: 0.7.19
+Version: 0.7.19.post1
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu
@@ -39,12 +39,14 @@ Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Provides-Extra: train
 Requires-Dist: accelerate~=0.27; extra == "train"
+Provides-Extra: docs
+Requires-Dist: mkdocs-material; extra == "docs"
+Requires-Dist: mkdocs-click; extra == "docs"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
 Dynamic: description
 Dynamic: description-content-type
-Dynamic: license-file
 Dynamic: provides-extra
 Dynamic: requires-dist
 Dynamic: summary

{batchalign-0.7.19 → batchalign-0.7.19.post1}/batchalign.egg-info/requires.txt

@@ -29,5 +29,9 @@ openai-whisper>=20240930
 [dev]
 pytest
 
+[docs]
+mkdocs-material
+mkdocs-click
+
 [train]
 accelerate~=0.27

{batchalign-0.7.19 → batchalign-0.7.19.post1}/setup.py

@@ -65,6 +65,10 @@ setup(
         'train': [
             'accelerate~=0.27',
         ],
+        'docs': [
+            'mkdocs-material',
+            'mkdocs-click'
+        ],
         # 'speaker': [
         #     "nemo-toolkit~=1.21.0",
         #     "omegaconf~=2.3.0",

batchalign-0.7.19/batchalign/version

@@ -1,3 +0,0 @@
-0.7.19
-May 9th, 2025
-OpenAI Whisper Implementation