modelai4bharatIndicConformer multilingualSpeech-recognition checkpoint from AI4Bharat. Real deployment needs dialect, audio-quality and domain-specific error analysis.Speech recognitionIndian sourceIndian languagesHugging FaceMIT
modelai4bharatIndicTrans2 English → IndicDistilled translation checkpoint for English-to-Indic workflows. Review publisher access conditions and evaluate each target language separately.TranslationIndian sourceIndian languagesHugging FaceMIT
datasetai4bharatIN22 ConversationalConversational translation evaluation data from AI4Bharat. A possible starting point for language-specific evaluation after access and license review.TranslationIndian sourceIndian languagesHugging FaceCC BY 4.0
datasetai4bharatIN22 GeneralTranslation evaluation dataset covering Indian languages. Use its stated license and access conditions; keep evaluation data separate from training data.TranslationIndian sourceIndian languagesHugging FaceCC BY 4.0
evaluationAI4Bharatindic-llm-eval-suiteIndian-language evaluation tooling from AI4Bharat. Compare results only when task versions and evaluation settings agree.EvaluationIndic languageIndian sourceIndian languagesGitHubLicense unknown
toolAI4BharatIndicLLMSuiteA blueprint for creating Pretraining and Fine-Tuning datasets for Indic languagesDatasetsIndic languageIndian sourceIndian languagesGitHubMIT
frameworkAI4BharatIndicTrans2Translation models for 22 scheduled languages of IndiaTranslationIndic languageIndian sourceIndian languagesGitHubMIT