Voice control and question answering (22.10)

David Erik, Mollberg; Þorsteinn Daði, Gunnarsson

Voice control and question answering (22.10)

Clarin IS Repository

Authors: David Erik, Mollberg and Þorsteinn Daði, Gunnarsson

Item identifier: http://hdl.handle.net/20.500.12537/295

Date issued: 2022-09-27

Type: toolService

Language(s): Icelandic

Description: [English] The goal of this work package was to develop Kaldi recipes for voice control and question answering systems for Icelandic. We defined six tasks and either generated or gathered data for each, normalized the data and trained Kaldi language models. Included in this submission are six ASR language models, an acoustic model, the training data for the language model and all the code used to generate the data and create the models. For further information have a look at the file README.md. [Icelandic] Markmiðið með þessu verkefni var að búa til talgreiningar uppskriftir með Kalda fyrir raddskipanir og fyrirspurnir. Við skilgreindum sex verkefni og annaðhvort söfnuðum eða bjuggum til gögn fyrir hvert og eitt þeirra, undirbjuggum gögnin og þjálfuðum mállíkön. Í þessu safni er að finna sex sérhæfð mállíkön, hljóðlíkan, gögnin sem voru notuð til þess að búa til mállíkönin ásamt öllum kóða sem notaður var til þess að búa til gögnin og líkönin. Freakri upplýsingar má finna í skránni README.md.

Publisher: Tiro

Reykjavik University

Subject(s): Language models LM ASR Automatic Speech Recognition

Collection(s): Clarin IS

Show full item record

Files in this item

Download all files in item (665.2 MB)

This item is

Publicly Available

and licensed under:
Creative Commons - Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

Name: h10_voice_control_and_question_answering_models.tar.gz
Size: 665.19 MB
Format: application/gzip
MD5: 0c46937054c9be51aab226d25a3d95f7

Download file Preview

File Preview

code
- is-trivia-questions-M9
  - normalize.py2 kB
  - README.md2 kB
  - data_is-triva
    - notes.md623 B
    - is-trivia.tsv1 MB
    - final.txt662 kB
    - is-trivia.csv934 kB
    - is-trivia_sorted_filtered.tsv1 MB
    - is-trivia_sorted_filtered_cleaned.tsv1 MB
    - bad-norm-examples7 kB
  - data_spurning-is
    - final.txt746 kB
    - spuringar-is_norm.tsv834 kB
    - spuringar-is.csv799 kB
    - spuringar-is_norm_cleaned.tsv811 kB
- unit-conversion-M9
  - create_list_of_nums.py911 B
  - src
    - unitconversion
      - main.py9 kB
      - units.py18 kB
      - utils.py1 kB
      - __init__.py94 B
      - sentences.py8 kB
  - README.md1 kB
  - .gitignore3 kB
  - output
  - tests
    - test_ints.py1 MB
    - __pycache__
      - test_ints.cpython-38-pytest-7.1.3.pyc858 kB
  - setup.cfg623 B
  - pyproject.toml495 B
  - requirements.txt150 B
  - .gitattributes389 B
  - generate_numbers
    - fill_in_fractions11 kB
    - small_test.tsv1 kB
    - nums.py11 kB
    - fill_in.tsv17 kB
    - numbers.tsv346 kB
    - nums_1to122.tsv29 kB
  - input
    - ints_1_999999_first_1000_repeated.tsv134 B
    - fractions_0_100_repeated.tsv1 MB
    - ints_1.tsv419 B
    - integers_0_1000.tsv323 kB
    - main_numbers.tsv134 B
    - 21_int.tsv708 B
    - ints_1_10.tsv1 kB
    - ints_1_999999.tsv16 MB
    - fractions_1.tsv3 kB
    - .gitattributes48 B
    - fractions_17 kB
    - fractions_0.tsv2 kB
    - numbers.tsv132 B
    - integers.tsv323 kB
    - fractions_0_10.tsv25 kB
    - long_list_ints.tsv58 MB
  - LICENSE1 kB
  - currencies
    - curr.tsv0 B
    - currency.txt3 kB
    - out8 kB
    - script.py7 kB
    - Gjaldmiðlar - Sheet4.csv4 kB
    - out22 kB
- numi-M9
  - src
    - numi
      - main.py9 kB
      - handle_input.py5 kB
      - utils.py3 kB
      - __init__.py2 kB
  - README.md3 kB
  - .gitignore3 kB
  - tests
    - test_numi.py18 kB
    - todo_test_numi_decimals.py1 kB
    - __init__.py0 B
  - setup.cfg1 kB
  - requirements_dev.txt40 B
  - .gitlab-ci.yml2 kB
  - pyproject.toml485 B
  - .github
    - workflows
      - test.yaml583 B
  - LICENSE1 kB
- create_models-M9
  - README.md7 kB
  - .gitignore18 B
  - cmd.sh143 B
  - decode_and_score.sh1 kB
  - tools
    - tiro
      - speech
        v1alpha
        speech_pb2.py15 kB
        speech_pb2_grpc.py4 kB
        __pycache__
        speech_pb2.cpython-38.pyc7 kB
        speech_pb2_grpc.cpython-38.pyc3 kB
    - create_test_recs.py2 kB
    - run_g2p.py1 kB
    - start_TSC_server.sh1 kB
    - use_mic.sh460 B
    - recognize.py7 kB
    - decode_and_score.sh1 kB
    - get_words_from_lexicon.py433 B
    - filter.py372 B
    - to_lower.py96 B
  - requirements.txt63 B
  - run.sh8 kB
  - .gitattributes448 B
  - path.sh439 B
  - setup.sh146 B
  - LICENSE1 kB
- fstring2fst-M9-3
  - .gitignore78 B
  - README.md1 kB
  - setup.py424 B
  - LICENSE.txt1 kB
  - example.py705 B
  - src
    - __init__.py0 B
    - fstring2fst.py2 kB
- lm-is-forms-M9
  - generate_data.sh387 B
  - collect_addresses.py5 kB
  - README.md1 kB
  - full_names.py568 B
  - phone_numbers.py199 B
  - generate_data.py1 kB
  - normalize_numbers.py2 kB
  - addresses.py540 B
  - raw
    - collect_names.sh626 B
    - prepare_addresses.sh660 B
    - INFO.md252 B
    - csv2tsv.py212 B
  - spell_number.py1 kB
  - kennitala.py501 B
data
- unit-conversion
  - lexicon.txt10 kB
  - train175 MB
  - test22 kB
- kennitolur
  - lexicon700 B
  - train19 MB
  - test14 kB
- trivia
  - lexicon1017 kB
  - train1 MB
  - test12 kB
- phone_numbers
  - lexicon861 B
  - train32 MB
  - test11 kB
- names
  - lexicon564 kB
  - train2 MB
  - test7 kB
- addresses
  - lexicon412 kB
  - train2 MB
  - test6 kB
- README.md9 kB
models
- unit-conversion
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops17 kB
    - wer196 B
    - per_spk2 kB
    - decode21 kB
    - per_utt81 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - phones
      - optional_silence.txt4 B
      - disambig.txt6 B
      - align_lexicon.txt19 kB
      - optional_silence.int2 B
      - disambig.int8 B
      - align_lexicon.int12 kB
      - word_boundary.txt2 kB
      - silence.csl21 B
      - word_boundary.int2 kB
      - optional_silence.csl2 B
    - words.txt4 kB
    - HCLG.fst21 MB
    - phones.txt1 kB
    - num_pdfs5 B
    - disambig_tid.int12 B
  - main.conf582 B
  - G.fst3 MB
  - tree1 MB
- kennitolur
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops3 kB
    - wer194 B
    - per_spk2 kB
    - decode12 kB
    - per_utt55 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - phones
      - optional_silence.txt4 B
      - disambig.txt6 B
      - align_lexicon.txt1 kB
      - optional_silence.int2 B
      - disambig.int8 B
      - align_lexicon.int810 B
      - word_boundary.txt1 kB
      - silence.csl21 B
      - word_boundary.int1 kB
      - optional_silence.csl2 B
    - words.txt332 B
    - HCLG.fst11 MB
    - phones.txt1 kB
    - num_pdfs5 B
    - disambig_tid.int12 B
  - main.conf582 B
  - G.fst1 MB
  - tree1 MB
- trivia
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops45 kB
    - wer191 B
    - per_spk2 kB
    - decode10 kB
    - per_utt46 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - phones
      - optional_silence.txt4 B
      - disambig.txt21 B
      - align_lexicon.txt1 MB
      - optional_silence.int2 B
      - disambig.int28 B
      - align_lexicon.int1 MB
      - word_boundary.txt2 kB
      - silence.csl21 B
      - word_boundary.int2 kB
      - optional_silence.csl2 B
    - words.txt511 kB
    - HCLG.fst90 MB
    - phones.txt2 kB
    - num_pdfs5 B
    - disambig_tid.int42 B
  - main.conf582 B
  - G.fst11 MB
  - tree1 MB
- phone_numbers
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops4 kB
    - wer197 B
    - per_spk2 kB
    - decode10 kB
    - per_utt46 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - phones
      - optional_silence.txt4 B
      - disambig.txt6 B
      - align_lexicon.txt1 kB
      - optional_silence.int2 B
      - disambig.int8 B
      - align_lexicon.int1003 B
      - word_boundary.txt1 kB
      - silence.csl21 B
      - word_boundary.int1 kB
      - optional_silence.csl2 B
    - words.txt405 B
    - HCLG.fst12 MB
    - phones.txt1 kB
    - num_pdfs5 B
    - disambig_tid.int12 B
  - main.conf582 B
  - G.fst2 MB
  - tree1 MB
- names
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops30 kB
    - wer186 B
    - per_spk2 kB
    - decode7 kB
    - per_utt31 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - phones
      - optional_silence.txt4 B
      - disambig.txt18 B
      - align_lexicon.txt1 MB
      - optional_silence.int2 B
      - disambig.int24 B
      - align_lexicon.int703 kB
      - word_boundary.txt2 kB
      - silence.csl21 B
      - word_boundary.int2 kB
      - optional_silence.csl2 B
    - words.txt265 kB
    - HCLG.fst95 MB
    - phones.txt2 kB
    - num_pdfs5 B
    - disambig_tid.int36 B
  - main.conf582 B
  - G.fst16 MB
  - tree1 MB
- addresses
  - frame_subsampling_factor1 B
  - phones.txt2 kB
  - results
    - ops17 kB
    - wer192 B
    - per_spk2 kB
    - decode5 kB
    - per_utt27 kB
  - ivector_extractor
    - final.ie18 MB
    - global_cmvn.stats1 kB
    - online_cmvn.conf108 B
    - final.mat43 kB
    - splice_opts35 B
    - final.dubm164 kB
  - final.mdl79 MB
  - conf
    - online_cmvn.conf108 B
    - splice.conf35 B
    - ivector_extractor.conf353 B
    - mfcc.conf669 B
  - graph
    - HCLGa.fst.640110 B
    - Ha.fst203 kB
    - phones
      - optional_silence.txt4 B
      - disambig.txt15 B
      - align_lexicon.txt769 kB
      - optional_silence.int2 B
      - disambig.int20 B
      - align_lexicon.int507 kB
      - word_boundary.txt2 kB
      - silence.csl21 B
      - word_boundary.int2 kB
      - optional_silence.csl2 B
    - words.txt195 kB
    - HCLG.fst17 MB
    - phones.txt2 kB
    - num_pdfs5 B
    - disambig_tid.int30 B
  - main.conf582 B
  - G.fst7 MB
  - tree1 MB

Name: README.md
Size: 9.66 KB
Format: Unknown
MD5: ab1b141e31c798ccfdcd82009d549568

Download file

Voice control and question answering (22.10)

Files in this item

Partners, Coordination, Funding

Repository

More