MAFIA (Match-Finder Aligner): A speech/text aligning tool (22.06)

Clarin IS Repository

Authors: Hernández Mena, Carlos Daniel

Item identifier: http://hdl.handle.net/20.500.12537/215

Date issued: 2022-05-30

Type: toolService

Language(s): Icelandic

Description: MAFIA is the acronym for Match-Finder Aligner. The MAFIA aligner is a software tool destined to automatically create ASR corpora out of speech files along with scripts reflecting what is spoken in such speech files. If the text of the scripts is not completely accurate, MAFIA will infer a transcription using automatic speech recognition.

MAFIA stendur fyrir "Match-Finder Aligner" eða samstæðufinnandi raðara á íslensku. MAFIA er hannað til að búa til talgreiningargögn á sjálfvirkan hátt úr upptökum og handriti sem endurspeglar innihald upptökunnar. Ef handritið er ekki fullkomlega rétt lesið mun MAFIA nota talgreiningu til að finna réttan texta.

Publisher: Language and Voice Lab (LVL), Reykjavík University

Subject(s): aligner segmenter speech recognition

Collection(s): Clarin IS

Show full item record

Files in this item

This item is

Publicly Available

and licensed under:
GNU General Public Licence, version 3

Name: MAFIA_v1.zip
Size: 27.73 MB
Format: application/zip
Description: MAFIA (Match-Finder Aligner)
MD5: a77d1ebfe23f3ee7426573f82d92b0f4

Download file Preview

File Preview

MAFIA_v1
- steps
  - find_word_matches.py-1 B
  - create_lm.py-1 B
  - segment_audio.py-1 B
  - clean_text.py-1 B
  - find_data_paths.py-1 B
  - create_segment_csvs.py-1 B
  - transcribe_segments.py-1 B
  - utils
    - matrix_handler.py-1 B
    - libcrocomad.py-1 B
    - csv_handler.py-1 B
    - match_finder.py-1 B
  - generate_results_file.py-1 B
  - dtw_aligment.py-1 B
- mafia.sh-1 B
- README-1 B
- models
  - QuartzNet15x1SEP-IS.nemo-1 B

Partners, Coordination, Funding

Repository

More

CLARIN-IS is fully supported by the Ministry of Culture and Business Affairs

Copyright (c) 2023. Arni Magnusson Institute for Icelandic Studies. All rights reserved.