BatchalignHK 0.7.21__tar.gz → 0.7.21.post2__tar.gz

{batchalignhk-0.7.21 → batchalignhk-0.7.21.post2}/BatchalignHK.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: BatchalignHK
-Version: 0.7.21
+Version: 0.7.21.post2
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.21 → batchalignhk-0.7.21.post2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: BatchalignHK
-Version: 0.7.21
+Version: 0.7.21.post2
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.21 → batchalignhk-0.7.21.post2}/batchalign/pipelines/asr/utils.py

@@ -168,7 +168,7 @@ def process_generation(output, lang="eng", utterance_engine=None):
         # which is ['word', [start_ms, end_ms]]. Yes, this would
         # involve multiplying by 1000 to s => ms
         words = [[i["value"], [round(i["ts"]*1000) if i.get("ts") != None else None,
-                                round(i["end_ts"]*1000) if i.get("end_ts") != None else None]] # the shape
+                               round(i["end_ts"]*1000) if i.get("end_ts") != None else None]] # the shape
                 for i in words # for each word
                     if i["value"].strip() != "" and
                     not re.match(r'<.*>', i["value"])] # if its text (i.e. not "pause")

{batchalignhk-0.7.21 → batchalignhk-0.7.21.post2}/batchalign/pipelines/morphosyntax/ud.py

@@ -769,11 +769,17 @@ def morphoanalyze(doc: Document, retokenize:bool, skipmultilang:bool, status_hoo
         else:
             config["processors"]["mwt"] = "default"
 
+    if "ja" in lang:
+        config["processors"]["tokenize"] = "combined"
+        config["processors"]["pos"] = "combined"
+        config["processors"]["lemma"] = "combined"
+        config["processors"]["depparse"] = "combined"
+
+
     configs = {}
     for l in lang:
         configs[l] = config.copy()
 
-
     if len(lang) > 1:
         nlp = stanza.MultilingualPipeline(
             lang_configs = configs,

batchalignhk-0.7.21.post2/batchalign/version

@@ -0,0 +1,3 @@
+0.7.21-post.2
+August 18, 2025
+ASR fixes

batchalignhk-0.7.21/batchalign/version

@@ -1,3 +0,0 @@
-0.7.21
-August 05, 2025
-FunAudio ASR