Datasets
Dataset
omnivoxtral-data
by viai957
OmniVoxtral Data This repository is the working-data backup of the OmniVoxtral project, a multilingual speech language model for Indic languages and English. The base model is Mistral-7B-v0.
No licenseautomatic-speech-recognition 262 0100K < n < 1M rows
published 4 Jun 2026
Preview
Couldn't infer the same data file format for all splits. Got {NamedSplit('train'): ('imagefolder', {}), NamedSplit('validation'): ('json', {}), NamedSplit('test'): ('webdataset', {})}
Cite this
No licenseCite
Tags
task_categories:automatic-speech-recognitiontask_categories:text-to-speechannotations_creators:machine-generatedmultilinguality:multilingualsource_datasets:google/fleurssource_datasets:ai4bharat/IndicVoicessource_datasets:originallanguage:aslanguage:bnlanguage:enlanguage:gulanguage:hilanguage:knlanguage:mllanguage:mrlanguage:nelanguage:orlanguage:palanguage:talanguage:telanguage:urlicense:othersize_categories:100K<n<1Marxiv:2205.12446arxiv:2403.01926arxiv:2410.00037arxiv:2212.04356arxiv:2305.13516region:usspeechspeech-language-modelindicmimineural-codec-tokensinterleaved-text-audiowhisper-pseudo-labelsresearch-onlyomnivoxtral