new utt2sub and sub2utt command

Mathias Quillot
1 parent d727acc21a
Showing 1 changed file with 72 additions and 1 deletions Side-by-side Diff
volia/masseffect.py
@@ -78,6 +78,48 @@
             core.data.write_line(",".join(splited), data_dict[key], out=f)
  
  
+def converter(file: str, outtype: str, outfile: str):
+    data = core.data.read_id_values(file)
+
+    with open(outfile, "w") as of:
+        for key in data:
+            splited = key.replace("\n", "").split(",")
+            masseffect_id = key.replace("\n", "")
+            kaldi_id = ",".join([splited[0], splited[1], splited[3]])
+            if outtype == "masseffect2kaldi":
+                of.write(f"{masseffect_id} {kaldi_id}\n")
+            elif outtype == "kaldi2masseffect":
+                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="...")
@@ -108,6 +150,31 @@
     parser_changelabels.add_argument("--outfile", required=True, type=str, help="Output file")
     parser_changelabels.set_defaults(which="changelabels")
  
+    # Create converter
+    parser_converter = subparsers.add_parser("converter", help="Create converter file")
+    parser_converter.add_argument("--file",
+                                  type=str,
+                                  required=True,
+                                  help="File with ids from which create converter.")
+    parser_converter.add_argument("--outtype", type=str, choices=["kaldi2masseffect", "masseffect2kaldi"])
+    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()
  
@@ -116,7 +183,10 @@
         "utt2char" : utt2char,
         "char2utt": char2utt,
         "wavscp": wavscp,
-        "changelabels": changelabels
+        "changelabels": changelabels,
+        "converter": converter,
+        "utt2sub": utt2sub,
+        "sub2utt": sub2utt
     })
  
     runner.run(args.which, args.__dict__, remove="which")
...	...	@@ -78,6 +78,48 @@
78	78	core.data.write_line(",".join(splited), data_dict[key], out=f)
79	79
80	80
	81	+def converter(file: str, outtype: str, outfile: str):
	82	+ data = core.data.read_id_values(file)
	83	+
	84	+ with open(outfile, "w") as of:
	85	+ for key in data:
	86	+ splited = key.replace("\n", "").split(",")
	87	+ masseffect_id = key.replace("\n", "")
	88	+ kaldi_id = ",".join([splited[0], splited[1], splited[3]])
	89	+ if outtype == "masseffect2kaldi":
	90	+ of.write(f"{masseffect_id} {kaldi_id}\n")
	91	+ elif outtype == "kaldi2masseffect":
	92	+ of.write(f"{kaldi_id} {masseffect_id}\n")
	93	+
	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	+
81	123	if __name__ == '__main__':
82	124	# Main parser
83	125	parser = argparse.ArgumentParser(description="...")
...	...	@@ -108,6 +150,31 @@
108	150	parser_changelabels.add_argument("--outfile", required=True, type=str, help="Output file")
109	151	parser_changelabels.set_defaults(which="changelabels")
110	152
	153	+ # Create converter
	154	+ parser_converter = subparsers.add_parser("converter", help="Create converter file")
	155	+ parser_converter.add_argument("--file",
	156	+ type=str,
	157	+ required=True,
	158	+ help="File with ids from which create converter.")
	159	+ parser_converter.add_argument("--outtype", type=str, choices=["kaldi2masseffect", "masseffect2kaldi"])
	160	+ parser_converter.add_argument("--outfile", type=str, required=True, help="")
	161	+ parser_converter.set_defaults(which="converter")
	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")
	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	+
111	178	# Parse
112	179	args = parser.parse_args()
113	180
...	...	@@ -116,7 +183,10 @@
116	183	"utt2char" : utt2char,
117	184	"char2utt": char2utt,
118	185	"wavscp": wavscp,
119		- "changelabels": changelabels
	186	+ "changelabels": changelabels,
	187	+ "converter": converter,
	188	+ "utt2sub": utt2sub,
	189	+ "sub2utt": sub2utt
120	190	})
121	191
122	192	runner.run(args.which, args.__dict__, remove="which")