Giter Club home page Giter Club logo

multipa's People

Contributors

ctaguchi avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar

multipa's Issues

Error: Could not find a version that satisfies the requirement torch==1.11.0

I just ran:

python3.11 -m pip install --upgrade pip
pip install -r requirements.txt

And the requirements fetch fails on torch mismatch, any ideas how to fix (I am no python expert yet):

¤ pip install -r requirements.txt
Collecting transformers@ git+https://github.com/huggingface/transformers@799cea64ac1029d66e9e58f18bc6f47892270723 (from -r requirements.txt (line 14))
  Cloning https://github.com/huggingface/transformers (to revision 799cea64ac1029d66e9e58f18bc6f47892270723) to /private/var/folders/8x/z26qdb3x465gmtqr6z4x8hgm0000gn/T/pip-install-7i6y5txl/transformers_e094f57325924970b06bcd4058e74de3
  Running command git clone --filter=blob:none --quiet https://github.com/huggingface/transformers /private/var/folders/8x/z26qdb3x465gmtqr6z4x8hgm0000gn/T/pip-install-7i6y5txl/transformers_e094f57325924970b06bcd4058e74de3
  Running command git rev-parse -q --verify 'sha^799cea64ac1029d66e9e58f18bc6f47892270723'
  Running command git fetch -q https://github.com/huggingface/transformers 799cea64ac1029d66e9e58f18bc6f47892270723
  Running command git checkout -q 799cea64ac1029d66e9e58f18bc6f47892270723
  Resolved https://github.com/huggingface/transformers to commit 799cea64ac1029d66e9e58f18bc6f47892270723
  Installing build dependencies ... done
  Getting requirements to build wheel ... done
  Installing backend dependencies ... done
  Preparing metadata (pyproject.toml) ... done
Collecting datasets==2.9.0 (from -r requirements.txt (line 1))
  Using cached datasets-2.9.0-py3-none-any.whl.metadata (19 kB)
Collecting epitran==1.24 (from -r requirements.txt (line 2))
  Using cached epitran-1.24-py2.py3-none-any.whl.metadata (34 kB)
Collecting ffmpeg-python==0.2.0 (from -r requirements.txt (line 3))
  Using cached ffmpeg_python-0.2.0-py3-none-any.whl.metadata (1.7 kB)
Collecting jiwer==2.5.1 (from -r requirements.txt (line 4))
  Using cached jiwer-2.5.1-py3-none-any.whl.metadata (11 kB)
Collecting librosa==0.9.2 (from -r requirements.txt (line 5))
  Using cached librosa-0.9.2-py3-none-any.whl.metadata (8.2 kB)
Collecting mecab-python3==1.0.6 (from -r requirements.txt (line 6))
  Using cached mecab_python3-1.0.6-cp311-cp311-macosx_10_9_universal2.whl.metadata (6.1 kB)
Collecting pandas==1.5.2 (from -r requirements.txt (line 7))
  Using cached pandas-1.5.2-cp311-cp311-macosx_11_0_arm64.whl.metadata (11 kB)
Collecting panphon==0.20.0 (from -r requirements.txt (line 8))
  Using cached panphon-0.20.0-py2.py3-none-any.whl.metadata (15 kB)
Collecting pyarrow==8.0.0 (from -r requirements.txt (line 9))
  Using cached pyarrow-8.0.0.tar.gz (846 kB)
  Installing build dependencies ... done
  Getting requirements to build wheel ... done
  Preparing metadata (pyproject.toml) ... done
Requirement already satisfied: pydub==0.25.1 in /opt/homebrew/lib/python3.11/site-packages (from -r requirements.txt (line 10)) (0.25.1)
Collecting romkan==0.2.1 (from -r requirements.txt (line 11))
  Using cached romkan-0.2.1.tar.gz (10 kB)
  Preparing metadata (setup.py) ... done
ERROR: Could not find a version that satisfies the requirement torch==1.11.0 (from versions: 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2)
ERROR: No matching distribution found for torch==1.11.0

Thanks! Looking forward to testing this out and following along your paper.

Summary of How Multipa Works?

Hi, I reached out to you via email just now FYI, just introducing myself basically, but wanted to follow up here.

What is required to support all IPA characteristics across all languages?

I am interested in languages with interesting phonemic patterns. Basically I just would like to cover all the ground of the pronunciation possibilities, in as short a word list as possible.

  • Tones (Chinese, Burmese, Yoruba, etc.). Should include single tone words (Yoruba has a lot), and shifting tones (Chinese has 2-tone vowels, Burmese, Thai, and Tibetan have 2-tone and also 3-tone vowels if I remember correctly).
  • Clicks (Xhosa is a good candidate, I do a lot with this language). Should cover clicks with and without aspiration, and include all possible clicks. One click is only found in !Xoo, so need at least one word from that.
  • Words with palatalization (Russian has a lot of words with this).
  • Glottalization (Arabic has this)
  • Velarization (Irish has this)
  • Pharyngealization (Arabic has this, know of any others?)
  • Labialization (Not sure..., some Native American languages I think)
  • Aspiration (Hindi or Sanskrit words)
  • All vowels (English has many as a good start, but then check out Danish, Swedish, and other European languages with lots of vowels)
  • All consonants (English is a good start, Hindi/Sanskrit for retroflex consonants, Navajo for that ɬ sound, Zulu I think for ɮ, french for "back r" ʁ, Hebrew for harsh χ).

How far is multipa from supporting all these features? What would it take (what work is there to be done to support these features)? Perhaps I could help in some way.

Does this work from Text to IPA, or Audio to IPA?

I can't tell but looking briefly at the code it appears this is a text transcription 2 IPA library, not an audio to IPA, is that correcft? Do you have an audio to IPA library elsewhere if correct? If not, what am I missing? (I am looking at things like finnish_to_ipa.py for example).

Basically, how do I do audio to IPA, given your paper 😁 ❤️ 🚀

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.