MMS-1B-all (Medumba adapter)

Medumba(byv)ASR / Wav2Vec2 + CTCWav2Vec2-XLS-RWav2Vec2ForCTC

View Source

Model Info

Language

Medumba (byv)

Architecture

Wav2Vec2ForCTC

Institution

Meta AI - Vineel Pratap et al.

Last Updated

21 May 2026

Intended Use Cases

Multilingual ASR (1,162 langs)

License & Access

SPDX identifierCC-BY-NC 4.0

Open with restrictions — attribution required; some commercial or derivative use limited.

Provenance

Source URL

https://huggingface.co/facebook/mms-1b-all

Fetch Timestamp

21 May 2026

Submitter

Meta AI - Vineel Pratap et al.

Full Record

23 of 25 fields stated

Every column the registry tracks. Values marked derivedwere filled by a rule or read from the source's text rather than stated in its metadata; the rule or the quoted sentence is shown under the value.

Registry columns (17/18)

iso code: byv
Model Name: MMS-1B-all (Medumba adapter)
Where is it hosted: HuggingFace
Link: https://huggingface.co/facebook/mms-1b-all
Year: 2023
Creator/Author(s): Meta AI - Vineel Pratap et al.
Paper Reference: Pratap et al. 2023 (arXiv:2305.13516)
NLP Task: ASR / Wav2Vec2 + CTC
Architecture: Wav2Vec2ForCTCderivedHugging Face config: architectures / model_type
Base Model: Wav2Vec2-XLS-R
Fine-tuned?: Yes
License: CC-BY-NC 4.0
Open Access: Open
Has Model Card: Yes
SOTA Tasks: Multilingual ASR (1,162 langs)
Verified By: Not stated
Date Indexed: 2026-05-21
Comment: Medumba adapter

Readiness (LRS) (6/7)

model id: mms-1b-all-medumba-adapter-byv
Model Type: ASRderivedHugging Face card: pipeline_tag
Training Data Sources: MMS-lab + FLEURS + CV + VP + MLS
Benchmark Used: MMS ASR Benchmark (Biafada adapter)derivedregistry benchmark records for this model
Benchmark Score: ASR WER for Biafada
Benchmark Date: 2023
Status: Not stated