Commit 16251ebfaabfb033919dc32d452a81f55bc71e40
1 parent
d0b2f9b221
Exists in
master
Fixed issue: adding \n at the end of the writing line.
Showing 1 changed file with 2 additions and 2 deletions Inline Diff
volia/masseffect.py
| 1 | import argparse | 1 | import argparse |
| 2 | from os import path | 2 | from os import path |
| 3 | import core.data | 3 | import core.data |
| 4 | from utils import SubCommandRunner | 4 | from utils import SubCommandRunner |
| 5 | import os | 5 | import os |
| 6 | 6 | ||
| 7 | 7 | ||
| 8 | def utt2char(features: str, outfile: str): | 8 | def utt2char(features: str, outfile: str): |
| 9 | """Allow the user to generate utt2char file from masseffect features file. | 9 | """Allow the user to generate utt2char file from masseffect features file. |
| 10 | 10 | ||
| 11 | TODO: Don't forget to manage two cases: one with old ids, and an other with | 11 | TODO: Don't forget to manage two cases: one with old ids, and an other with |
| 12 | new ones. | 12 | new ones. |
| 13 | 13 | ||
| 14 | Args: | 14 | Args: |
| 15 | features (str): [description] | 15 | features (str): [description] |
| 16 | outfile (str): [description] | 16 | outfile (str): [description] |
| 17 | """ | 17 | """ |
| 18 | data = core.data.read_features(features) | 18 | data = core.data.read_features(features) |
| 19 | keys = list(data.keys()) | 19 | keys = list(data.keys()) |
| 20 | 20 | ||
| 21 | with open(outfile, "w") as f: | 21 | with open(outfile, "w") as f: |
| 22 | for key in keys: | 22 | for key in keys: |
| 23 | splited = key.replace("\n", "").split(",") | 23 | splited = key.replace("\n", "").split(",") |
| 24 | character = splited[1] | 24 | character = splited[1] |
| 25 | f.write(",".join(splited) + " " + character + "\n") | 25 | f.write(",".join(splited) + " " + character + "\n") |
| 26 | 26 | ||
| 27 | 27 | ||
| 28 | def char2utt(features: str, outfile: str): | 28 | def char2utt(features: str, outfile: str): |
| 29 | raise Exception("Not implemented yet") | 29 | raise Exception("Not implemented yet") |
| 30 | pass | 30 | pass |
| 31 | 31 | ||
| 32 | 32 | ||
| 33 | def wavscp(datadir: str, outfile: str): | 33 | def wavscp(datadir: str, outfile: str): |
| 34 | """Generate the masseffect wav scp file from the directories. | 34 | """Generate the masseffect wav scp file from the directories. |
| 35 | 35 | ||
| 36 | Args: | 36 | Args: |
| 37 | datadir (str): oath of the data directory where "audio_en-us" and "audio_fr-fr" are available | 37 | datadir (str): oath of the data directory where "audio_en-us" and "audio_fr-fr" are available |
| 38 | outfile (str): path of the wav scp output file | 38 | outfile (str): path of the wav scp output file |
| 39 | 39 | ||
| 40 | Raises: | 40 | Raises: |
| 41 | Exception: if one of the directory is not available | 41 | Exception: if one of the directory is not available |
| 42 | """ | 42 | """ |
| 43 | en_us_dir = os.path.join(datadir, "audio_en-us") | 43 | en_us_dir = os.path.join(datadir, "audio_en-us") |
| 44 | fr_fr_dir = os.path.join(datadir, "audio_fr-fr") | 44 | fr_fr_dir = os.path.join(datadir, "audio_fr-fr") |
| 45 | 45 | ||
| 46 | if (not os.path.isdir(en_us_dir)) or (not os.path.isdir(fr_fr_dir)): | 46 | if (not os.path.isdir(en_us_dir)) or (not os.path.isdir(fr_fr_dir)): |
| 47 | raise Exception("Directory audio_en-us or audio_fr-fr does not exist") | 47 | raise Exception("Directory audio_en-us or audio_fr-fr does not exist") |
| 48 | 48 | ||
| 49 | _,_,filenames_en=next(os.walk(en_us_dir)) | 49 | _,_,filenames_en=next(os.walk(en_us_dir)) |
| 50 | # filenames_en = [ os.path.join(en_us_dir, f) for f in filenames_en ] | 50 | # filenames_en = [ os.path.join(en_us_dir, f) for f in filenames_en ] |
| 51 | dir_en = [ en_us_dir for f in filenames_en ] | 51 | dir_en = [ en_us_dir for f in filenames_en ] |
| 52 | _,_,filenames_fr=next(os.walk(fr_fr_dir)) | 52 | _,_,filenames_fr=next(os.walk(fr_fr_dir)) |
| 53 | dir_fr = [ fr_fr_dir for f in filenames_fr ] | 53 | dir_fr = [ fr_fr_dir for f in filenames_fr ] |
| 54 | # filenames_fr = [ os.path.join(fr_fr_dir, f) for f in filenames_fr ] | 54 | # filenames_fr = [ os.path.join(fr_fr_dir, f) for f in filenames_fr ] |
| 55 | 55 | ||
| 56 | directories = dir_en + dir_fr | 56 | directories = dir_en + dir_fr |
| 57 | filenames = filenames_en + filenames_fr | 57 | filenames = filenames_en + filenames_fr |
| 58 | 58 | ||
| 59 | with open(outfile, "w") as f: | 59 | with open(outfile, "w") as f: |
| 60 | for i, fn in enumerate(filenames): | 60 | for i, fn in enumerate(filenames): |
| 61 | splited = fn.split(".")[0].split(",") | 61 | splited = fn.split(".")[0].split(",") |
| 62 | lang = splited[0] | 62 | lang = splited[0] |
| 63 | character = splited[1] | 63 | character = splited[1] |
| 64 | record_id = splited[3] | 64 | record_id = splited[3] |
| 65 | path = os.path.join(directories[i], fn) | 65 | path = os.path.join(directories[i], fn) |
| 66 | f.write(f"{lang},{character},{record_id} {path}\n") | 66 | f.write(f"{lang},{character},{record_id} {path}\n") |
| 67 | 67 | ||
| 68 | 68 | ||
| 69 | def changelabels(source: str, labels: str, outfile: str): | 69 | def changelabels(source: str, labels: str, outfile: str): |
| 70 | data_dict = core.data.read_id_values(source) | 70 | data_dict = core.data.read_id_values(source) |
| 71 | labels_dict = core.data.read_labels(labels) | 71 | labels_dict = core.data.read_labels(labels) |
| 72 | keys = list(data_dict.keys()) | 72 | keys = list(data_dict.keys()) |
| 73 | 73 | ||
| 74 | with open(outfile, "w") as f: | 74 | with open(outfile, "w") as f: |
| 75 | for key in keys: | 75 | for key in keys: |
| 76 | splited = key.split(",") | 76 | splited = key.split(",") |
| 77 | splited[1] = labels_dict[key][0] | 77 | splited[1] = labels_dict[key][0] |
| 78 | core.data.write_line(",".join(splited), data_dict[key], out=f) | 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): | 81 | def converter(file: str, outtype: str, outfile: str): |
| 82 | data = core.data.read_id_values(file) | 82 | data = core.data.read_id_values(file) |
| 83 | 83 | ||
| 84 | with open(outfile, "w") as of: | 84 | with open(outfile, "w") as of: |
| 85 | for key in data: | 85 | for key in data: |
| 86 | splited = key.replace("\n", "").split(",") | 86 | splited = key.replace("\n", "").split(",") |
| 87 | masseffect_id = key.replace("\n", "") | 87 | masseffect_id = key.replace("\n", "") |
| 88 | kaldi_id = ",".join([splited[0], splited[1], splited[3]]) | 88 | kaldi_id = ",".join([splited[0], splited[1], splited[3]]) |
| 89 | if outtype == "masseffect2kaldi": | 89 | if outtype == "masseffect2kaldi": |
| 90 | of.write(f"{masseffect_id} {kaldi_id}\n") | 90 | of.write(f"{masseffect_id} {kaldi_id}\n") |
| 91 | elif outtype == "kaldi2masseffect": | 91 | elif outtype == "kaldi2masseffect": |
| 92 | of.write(f"{kaldi_id} {masseffect_id}\n") | 92 | of.write(f"{kaldi_id} {masseffect_id}\n") |
| 93 | 93 | ||
| 94 | 94 | ||
| 95 | def utt2sub(file: str, subfile: str, outfile: str): | 95 | def utt2sub(file: str, subfile: str, outfile: str): |
| 96 | data = core.data.read_id_values(file) | 96 | data = core.data.read_id_values(file) |
| 97 | keys = [key for key in data] | 97 | keys = [key for key in data] |
| 98 | 98 | ||
| 99 | data_sub = core.data.read_id_values(subfile) | 99 | data_sub = core.data.read_id_values(subfile) |
| 100 | keys_sub = [key for key in data_sub] | 100 | keys_sub = [key for key in data_sub] |
| 101 | 101 | ||
| 102 | with open(outfile, "w") as of: | 102 | with open(outfile, "w") as of: |
| 103 | for key in keys: | 103 | for key in keys: |
| 104 | subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)] | 104 | subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)] |
| 105 | subkeys_str = " ".join(subkeys) | 105 | subkeys_str = " ".join(subkeys) |
| 106 | of.write(f"{key} {subkeys_str}") | 106 | of.write(f"{key} {subkeys_str}\n") |
| 107 | 107 | ||
| 108 | 108 | ||
| 109 | def sub2utt(file: str, subfile: str, outfile: str): | 109 | def sub2utt(file: str, subfile: str, outfile: str): |
| 110 | data = core.data.read_id_values(file) | 110 | data = core.data.read_id_values(file) |
| 111 | keys = [key for key in data] | 111 | keys = [key for key in data] |
| 112 | 112 | ||
| 113 | data_sub = core.data.read_id_values(subfile) | 113 | data_sub = core.data.read_id_values(subfile) |
| 114 | keys_sub = [key for key in data_sub] | 114 | keys_sub = [key for key in data_sub] |
| 115 | 115 | ||
| 116 | with open(outfile, "w") as of: | 116 | with open(outfile, "w") as of: |
| 117 | for key in keys: | 117 | for key in keys: |
| 118 | subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)] | 118 | subkeys = [subkey for subkey in keys_sub if subkey.startswith(key)] |
| 119 | for subkey in subkeys: | 119 | for subkey in subkeys: |
| 120 | of.write(f"{subkey} {key}") | 120 | of.write(f"{subkey} {key}\n") |
| 121 | 121 | ||
| 122 | 122 | ||
| 123 | if __name__ == '__main__': | 123 | if __name__ == '__main__': |
| 124 | # Main parser | 124 | # Main parser |
| 125 | parser = argparse.ArgumentParser(description="...") | 125 | parser = argparse.ArgumentParser(description="...") |
| 126 | subparsers = parser.add_subparsers(title="action") | 126 | subparsers = parser.add_subparsers(title="action") |
| 127 | 127 | ||
| 128 | # utt2char | 128 | # utt2char |
| 129 | parser_utt2char = subparsers.add_parser("utt2char", help="generate utt2char file") | 129 | parser_utt2char = subparsers.add_parser("utt2char", help="generate utt2char file") |
| 130 | parser_utt2char.add_argument("--features", type=str, help="features file") | 130 | parser_utt2char.add_argument("--features", type=str, help="features file") |
| 131 | parser_utt2char.add_argument("--outfile", type=str, help="output file") | 131 | parser_utt2char.add_argument("--outfile", type=str, help="output file") |
| 132 | parser_utt2char.set_defaults(which="utt2char") | 132 | parser_utt2char.set_defaults(which="utt2char") |
| 133 | 133 | ||
| 134 | # char2utt | 134 | # char2utt |
| 135 | parser_char2utt = subparsers.add_parser("char2utt", help="generate char2utt file") | 135 | parser_char2utt = subparsers.add_parser("char2utt", help="generate char2utt file") |
| 136 | parser_char2utt.add_argument("--features", type=str, help="features file") | 136 | parser_char2utt.add_argument("--features", type=str, help="features file") |
| 137 | parser_char2utt.add_argument("--outfile", type=str, help="output file") | 137 | parser_char2utt.add_argument("--outfile", type=str, help="output file") |
| 138 | parser_char2utt.set_defaults(which="char2utt") | 138 | parser_char2utt.set_defaults(which="char2utt") |
| 139 | 139 | ||
| 140 | # wavscp | 140 | # wavscp |
| 141 | parser_wavscp = subparsers.add_parser("wavscp", help="generate wav scp file") | 141 | parser_wavscp = subparsers.add_parser("wavscp", help="generate wav scp file") |
| 142 | parser_wavscp.add_argument("--datadir", required=True, help="data directory of masseffect") | 142 | parser_wavscp.add_argument("--datadir", required=True, help="data directory of masseffect") |
| 143 | parser_wavscp.add_argument("--outfile", default="wav.scp", help="wav.scp output file") | 143 | parser_wavscp.add_argument("--outfile", default="wav.scp", help="wav.scp output file") |
| 144 | parser_wavscp.set_defaults(which="wavscp") | 144 | parser_wavscp.set_defaults(which="wavscp") |
| 145 | 145 | ||
| 146 | # Change labels | 146 | # Change labels |
| 147 | parser_changelabels = subparsers.add_parser("changelabels", help="...") | 147 | parser_changelabels = subparsers.add_parser("changelabels", help="...") |
| 148 | parser_changelabels.add_argument("--source", required=True, type=str, help="source file where we want to change ids.") | 148 | parser_changelabels.add_argument("--source", required=True, type=str, help="source file where we want to change ids.") |
| 149 | parser_changelabels.add_argument("--labels", required=True, type=str, help="file with labels") | 149 | parser_changelabels.add_argument("--labels", required=True, type=str, help="file with labels") |
| 150 | parser_changelabels.add_argument("--outfile", required=True, type=str, help="Output file") | 150 | parser_changelabels.add_argument("--outfile", required=True, type=str, help="Output file") |
| 151 | parser_changelabels.set_defaults(which="changelabels") | 151 | parser_changelabels.set_defaults(which="changelabels") |
| 152 | 152 | ||
| 153 | # Create converter | 153 | # Create converter |
| 154 | parser_converter = subparsers.add_parser("converter", help="Create converter file") | 154 | parser_converter = subparsers.add_parser("converter", help="Create converter file") |
| 155 | parser_converter.add_argument("--file", | 155 | parser_converter.add_argument("--file", |
| 156 | type=str, | 156 | type=str, |
| 157 | required=True, | 157 | required=True, |
| 158 | help="File with ids from which create converter.") | 158 | help="File with ids from which create converter.") |
| 159 | parser_converter.add_argument("--outtype", type=str, choices=["kaldi2masseffect", "masseffect2kaldi"]) | 159 | parser_converter.add_argument("--outtype", type=str, choices=["kaldi2masseffect", "masseffect2kaldi"]) |
| 160 | parser_converter.add_argument("--outfile", type=str, required=True, help="") | 160 | parser_converter.add_argument("--outfile", type=str, required=True, help="") |
| 161 | parser_converter.set_defaults(which="converter") | 161 | parser_converter.set_defaults(which="converter") |
| 162 | 162 | ||
| 163 | # Create utt2sub | 163 | # Create utt2sub |
| 164 | parser_utt2sub = subparsers.add_parser("utt2sub", help="generate utt2sub file") | 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") | 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") | 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") | 167 | parser_utt2sub.add_argument("--outfile", required=True, type=str, help="output file") |
| 168 | parser_utt2sub.set_defaults(which="utt2sub") | 168 | parser_utt2sub.set_defaults(which="utt2sub") |
| 169 | 169 | ||
| 170 | # Create sub2utt | 170 | # Create sub2utt |
| 171 | parser_sub2utt = subparsers.add_parser("sub2utt", help="generate sub2utt file") | 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") | 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") | 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") | 174 | parser_sub2utt.add_argument("--outfile", required=True, type=str, help="output file") |
| 175 | parser_sub2utt.set_defaults(which="sub2utt") | 175 | parser_sub2utt.set_defaults(which="sub2utt") |
| 176 | 176 | ||
| 177 | 177 | ||
| 178 | # Parse | 178 | # Parse |
| 179 | args = parser.parse_args() | 179 | args = parser.parse_args() |
| 180 | 180 | ||
| 181 | # Run commands | 181 | # Run commands |
| 182 | runner = SubCommandRunner({ | 182 | runner = SubCommandRunner({ |
| 183 | "utt2char" : utt2char, | 183 | "utt2char" : utt2char, |
| 184 | "char2utt": char2utt, | 184 | "char2utt": char2utt, |
| 185 | "wavscp": wavscp, | 185 | "wavscp": wavscp, |
| 186 | "changelabels": changelabels, | 186 | "changelabels": changelabels, |
| 187 | "converter": converter, | 187 | "converter": converter, |
| 188 | "utt2sub": utt2sub, | 188 | "utt2sub": utt2sub, |
| 189 | "sub2utt": sub2utt | 189 | "sub2utt": sub2utt |
| 190 | }) | 190 | }) |
| 191 | 191 | ||
| 192 | runner.run(args.which, args.__dict__, remove="which") | 192 | runner.run(args.which, args.__dict__, remove="which") |
| 193 | 193 | ||
| 194 | 194 |