Adding utt2sub and sub2utt commands

quillotm
1 parent 44433f16b4
Showing 1 changed file with 45 additions and 1 deletions Side-by-side Diff
volia/masseffect.py
@@ -92,6 +92,34 @@
                 of.write(f"{kaldi_id} {masseffect_id}\n")
  
  
+def utt2sub(self, file: str, subfile: str, outfile: str):
+    data = core.data.read_id_values(file)
+    keys = [key for key in data]
+
+    data_sub = core.data.read_id_values(subfile)
+    keys_sub = [key for key in data_sub]
+
+    with open(outfile) as of:
+        for key in keys:
+            subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)]
+            subkeys_str = " ".join(subkeys)
+            of.write(f"{key} {subkeys_str}")
+
+
+def sub2utt(self, file: str, subfile: str, outfile: str):
+    data = core.data.read_id_values(file)
+    keys = [key for key in data]
+
+    data_sub = core.data.read_id_values(subfile)
+    keys_sub = [key for key in data_sub]
+
+    with open(outfile) as of:
+        for key in keys:
+            subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)]
+            for subkey in subkeys:
+                of.write(f"{subkey} {key}")
+
+
 if __name__ == '__main__':
     # Main parser
     parser = argparse.ArgumentParser(description="...")
  
@@ -132,7 +160,21 @@
     parser_converter.add_argument("--outfile", type=str, required=True, help="")
     parser_converter.set_defaults(which="converter")
  
+    # Create utt2sub
+    parser_utt2sub = subparsers.add_parser("utt2sub", help="generate utt2sub file")
+    parser_utt2sub.add_argument("--file", required=True, type=str, help="features, list or labels file with normal ids")
+    parser_utt2sub.add_argument("--subfile", required=True, type=str, help="features, list or labels file with sub ids")
+    parser_utt2sub.add_argument("--outfile", required=True, type=str, help="output file")
+    parser_utt2sub.set_defaults(which="utt2sub")
  
+    # Create sub2utt
+    parser_sub2utt = subparsers.add_parser("sub2utt", help="generate sub2utt file")
+    parser_sub2utt.add_argument("--file", required=True, type=str, help="features, list or labels file with normal ids")
+    parser_sub2utt.add_argument("--subfile", required=True, type=str, help="features, list or labels file sub ids")
+    parser_sub2utt.add_argument("--outfile", required=True, type=str, help="output file")
+    parser_sub2utt.set_defaults(which="sub2utt")
+
+
     # Parse
     args = parser.parse_args()
  
@@ -142,7 +184,9 @@
         "char2utt": char2utt,
         "wavscp": wavscp,
         "changelabels": changelabels,
-        "converter": converter
+        "converter": converter,
+        "utt2sub": utt2sub,
+        "sub2utt": sub2utt
     })
  
     runner.run(args.which, args.__dict__, remove="which")
...	...	@@ -92,6 +92,34 @@
92	92	of.write(f"{kaldi_id} {masseffect_id}\n")
93	93
94	94
	95	+def utt2sub(self, file: str, subfile: str, outfile: str):
	96	+ data = core.data.read_id_values(file)
	97	+ keys = [key for key in data]
	98	+
	99	+ data_sub = core.data.read_id_values(subfile)
	100	+ keys_sub = [key for key in data_sub]
	101	+
	102	+ with open(outfile) as of:
	103	+ for key in keys:
	104	+ subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)]
	105	+ subkeys_str = " ".join(subkeys)
	106	+ of.write(f"{key} {subkeys_str}")
	107	+
	108	+
	109	+def sub2utt(self, file: str, subfile: str, outfile: str):
	110	+ data = core.data.read_id_values(file)
	111	+ keys = [key for key in data]
	112	+
	113	+ data_sub = core.data.read_id_values(subfile)
	114	+ keys_sub = [key for key in data_sub]
	115	+
	116	+ with open(outfile) as of:
	117	+ for key in keys:
	118	+ subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)]
	119	+ for subkey in subkeys:
	120	+ of.write(f"{subkey} {key}")
	121	+
	122	+
95	123	if __name__ == '__main__':
96	124	# Main parser
97	125	parser = argparse.ArgumentParser(description="...")
98	126
...	...	@@ -132,7 +160,21 @@
132	160	parser_converter.add_argument("--outfile", type=str, required=True, help="")
133	161	parser_converter.set_defaults(which="converter")
134	162
	163	+ # Create utt2sub
	164	+ parser_utt2sub = subparsers.add_parser("utt2sub", help="generate utt2sub file")
	165	+ parser_utt2sub.add_argument("--file", required=True, type=str, help="features, list or labels file with normal ids")
	166	+ parser_utt2sub.add_argument("--subfile", required=True, type=str, help="features, list or labels file with sub ids")
	167	+ parser_utt2sub.add_argument("--outfile", required=True, type=str, help="output file")
	168	+ parser_utt2sub.set_defaults(which="utt2sub")
135	169
	170	+ # Create sub2utt
	171	+ parser_sub2utt = subparsers.add_parser("sub2utt", help="generate sub2utt file")
	172	+ parser_sub2utt.add_argument("--file", required=True, type=str, help="features, list or labels file with normal ids")
	173	+ parser_sub2utt.add_argument("--subfile", required=True, type=str, help="features, list or labels file sub ids")
	174	+ parser_sub2utt.add_argument("--outfile", required=True, type=str, help="output file")
	175	+ parser_sub2utt.set_defaults(which="sub2utt")
	176	+
	177	+
136	178	# Parse
137	179	args = parser.parse_args()
138	180
...	...	@@ -142,7 +184,9 @@
142	184	"char2utt": char2utt,
143	185	"wavscp": wavscp,
144	186	"changelabels": changelabels,
145		- "converter": converter
	187	+ "converter": converter,
	188	+ "utt2sub": utt2sub,
	189	+ "sub2utt": sub2utt
146	190	})
147	191
148	192	runner.run(args.which, args.__dict__, remove="which")