Hugging Face's logo

utter-project
/
mHuBERT-147

Feature Extraction
Transformers
PyTorch
Safetensors
120 languages
hubert
Inference Endpoints
Model card Files Community
6
mHuBERT-147 / README.md
mzboito's picture
mzboito
Update README.md
8ba83f6 verified
preview code
|
raw
history blame
No virus
2.83 kB
metadata
license: cc-by-nc-4.0
language:
  - ab
  - af
  - am
  - ar
  - as
  - az
  - ba
  - be
  - bn
  - bo
  - bs
  - br
  - bg
  - ca
  - cs
  - cv
  - cy
  - da
  - de
  - dv
  - el
  - en
  - eo
  - et
  - eu
  - ee
  - fo
  - fa
  - tl
  - fi
  - fr
  - fy
  - ga
  - gl
  - gv
  - gn
  - gu
  - ht
  - ha
  - he
  - hi
  - hr
  - hu
  - hy
  - ig
  - ia
  - id
  - is
  - it
  - jv
  - ja
  - kn
  - ka
  - kk
  - km
  - rw
  - ky
  - ku
  - ko
  - lo
  - la
  - lv
  - ln
  - lt
  - lb
  - lg
  - ml
  - mr
  - mk
  - mg
  - mt
  - mn
  - mi
  - ms
  - my
  - ne
  - nl
  - nn
  - 'no'
  - oc
  - or
  - pa
  - pl
  - pt
  - ps
  - ro
  - ru
  - sa
  - si
  - sl
  - sk
  - sn
  - sd
  - so
  - st
  - es
  - sq
  - sc
  - sr
  - su
  - sw
  - sv
  - ta
  - tt
  - te
  - tg
  - th
  - tn
  - tk
  - tr
  - tw
  - ug
  - uk
  - ur
  - uz
  - vi
  - xh
  - yi
  - yo
  - zh

mHuBERT-147 models

mHuBERT-147 are multilingual general-purpose HuBERT models trained on 90K hours of open-license data in 147 languages.

This repository contains:

  • Fairseq checkpoint (original);
  • HuggingFace checkpoint;
  • Faiss index for continuous pre-training (OPQ16_64,IVF1000_HNSW32,PQ16x4fsr).

Citing 

[PAPER GOES HERE]
'''

# Other information

**Languages present not indexed by Huggingface:** Asturian (ast), Basaa (bas), Cebuano (ceb), Central Kurdish/Sorani (ckb), Hakha Chin (cnh), Hawaiian (haw), Upper Sorbian (hsb) Kabyle (kab), Moksha (mdf), Meadow Mari (mhr), Hill Mari (mrj), Erzya (myv), Taiwanese Hokkien (nan-tw), Sursilvan (rm-sursilv), Vallader (rm-vallader), Sakha (sah), Santali (sat), Scots (sco), Saraiki (skr), Tigre (tig), Tok Pisin (tpi), Akwapen Twi (tw-akuapem), Asante Twi (tw-asante), Votic (vot), Waray (war), Cantonese (yue).

**Manifest list:** check manifest/

**Datasets:** For ASR/ST/TTS datasets, only train set is used.
* [Aishell](https://www.openslr.org/33/) and [AISHELL-3](https://www.openslr.org/93/)
* [BibleTTS](https://www.openslr.org/129/)
* [ClovaCall](https://github.com/clovaai/ClovaCall)
* [CommonVoice v11](https://commonvoice.mozilla.org/en/datasets)
* Google TTS data: [Javanese](https://www.openslr.org/41/), [Khmer](https://www.openslr.org/42/), [Nepali](https://www.openslr.org/43/), [Sundanese](https://www.openslr.org/44/), [South African Languages](https://www.openslr.org/32/), [Bengali Languages](https://www.openslr.org/37/)  
* IISc-MILE: [Tamil](https://www.openslr.org/127/), [Kannada](https://www.openslr.org/126/)
* [Japanese Versatile Speech](https://sites.google.com/site/shinnosuketakamichi/research-topics/jvs_corpus) 
* [Kokoro](https://github.com/kaiidams/Kokoro-Speech-Dataset)
* [Kosp2e](https://github.com/warnikchow/kosp2e)
* Media Speech: [Turkish Only](https://www.openslr.org/108/)
* [Multilingual LibriSpeech](https://www.openslr.org/94/)
* [Samrómur](https://www.openslr.org/128/)
* [THCHS-30](https://www.openslr.org/18/) and [THUYG-20](https://www.openslr.org/22/)
* [VoxLingua107](https://bark.phon.ioc.ee/voxlingua107/)
* [VoxPopuli](https://github.com/facebookresearch/voxpopuli/)