Skip to content

[research] link to the XTREME-S paper #16519

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
anton-l merged 2 commits into from
Mar 31, 2022
Merged

[research] link to the XTREME-S paper #16519

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fe165fb
[research] link to the XTREME-S paper
anton-l Mar 31, 2022
98d94f5
Update examples/research_projects/xtreme-s/README.md
anton-l Mar 31, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions examples/research_projects/xtreme-s/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,9 @@ limitations under the License.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders for Speech (XTREME-S) benchmark is a benchmark designed to evaluate speech representations across languages, tasks, domains and data regimes. It covers XX typologically diverse languages and seven downstream tasks grouped in four families: speech recognition, translation, classification and retrieval.

XTREME-S covers speech recognition with Fleurs, Multilingual LibriSpeech (MLS) and VoxPopuli, speech translation with CoVoST-2, speech classification with LangID (Fleurs) and intent classification (MInds-14) and finally speech(-text) retrieval with Fleurs. Each of the tasks covers a subset of the 102 languages included in XTREME-S (shown here with their ISO 3166-1 codes): afr, amh, ara, asm, ast, azj, bel, ben, bos, cat, ceb, zho_simpl, zho_trad, ces, cym, dan, deu, ell, eng, spa, est, fas, ful, fin, tgl, fra, gle, glg, guj, hau, heb, hin, hrv, hun, hye, ind, ibo, isl, ita, jpn, jav, kat, kam, kea, kaz, khm, kan, kor, ckb, kir, ltz, lug, lin, lao, lit, luo, lav, mri, mkd, mal, mon, mar, msa, mlt, mya, nob, npi, nld, nso, nya, oci, orm, ory, pan, pol, pus, por, ron, rus, bul, snd, slk, slv, sna, som, srp, swe, swh, tam, tel, tgk, tha, tur, ukr, umb, urd, uzb, vie, wol, xho, yor and zul.
XTREME-S covers speech recognition with Fleurs, Multilingual LibriSpeech (MLS) and VoxPopuli, speech translation with CoVoST-2, speech classification with LangID (Fleurs) and intent classification (MInds-14) and finally speech(-text) retrieval with Fleurs. Each of the tasks covers a subset of the 102 languages included in XTREME-S (shown here with their ISO 3166-1 codes): afr, amh, ara, asm, ast, azj, bel, ben, bos, cat, ceb, ces, cmn, cym, dan, deu, ell, eng, spa, est, fas, ful, fin, tgl, fra, gle, glg, guj, hau, heb, hin, hrv, hun, hye, ind, ibo, isl, ita, jpn, jav, kat, kam, kea, kaz, khm, kan, kor, ckb, kir, ltz, lug, lin, lao, lit, luo, lav, mri, mkd, mal, mon, mar, msa, mlt, mya, nob, npi, nld, nso, nya, oci, orm, ory, pan, pol, pus, por, ron, rus, bul, snd, slk, slv, sna, som, srp, swe, swh, tam, tel, tgk, tha, tur, ukr, umb, urd, uzb, vie, wol, xho, yor, yue and zul.

Paper: `<TODO>`
Paper: [XTREME-S: Evaluating Cross-lingual Speech Representations](https://arxiv.org/abs/2203.10752)

Dataset: [https://huggingface.co/datasets/google/xtreme_s](https://huggingface.co/datasets/google/xtreme_s)

Expand Down