{"success":true,"data":{"author":"Google, with Makerere University, University of Ghana, Digital Umuganda, Media Trust, Loud and Clear, AIMS Senegal","bucketAccessType":"public","bucketName":"google/WaxalNLP","bucketPrefix":"","category":"Speech & Audio","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:17.794Z","description":"WAXAL is a large-scale multilingual speech corpus for African languages, released by Google with partner institutions and funded by Google and the Gates Foundation under an agreement to make it openly accessible. The ASR part covers 19 African languages with approximately 1,250 hours of transcribed natural speech from many voices, representing over 100 million speakers across 40 Sub-Saharan countries (including Acholi, Luganda, Akan, Ewe, Dagbani, Fula, Lingala, Shona, Malagasy, Amharic, Oromo, Tigrinya and Wolaytta). The TTS part covers 17 languages with over 180 hours of high-quality single-speaker recordings of phonetically balanced scripts (including Hausa, Igbo, Yoruba, Nigerian Pidgin, Kikuyu, Luo, Swahili, Bambara and Wolof). This repository holds data/ASR (19 language folders), data/ASR_v2 (speaker-disjoint splits) and data/TTS (18 folders) as Parquet shards, plus a Croissant metadata file (measured 2026-11-14; data/ASR/aka alone is 278 shards, about 57 GB; the Hub reports about 821 GB). Load with the datasets library per language. Licences are set per provider and are all open: University of Ghana data CC BY 4.0, all other providers CC BY-SA 4.0; keep the per-language licence when redistributing.","displayTitle":"WAXAL: multilingual African language speech corpus (ASR and TTS)","id":"8b8134f8-3490-5b08-8f03-40f5dfcea1a1","license":"CC-BY-SA-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/google/WaxalNLP","objectCountEstimate":2619,"probeStatus":"ok","referenceUrl":"https://huggingface.co/papers/2602.02734","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"google/WaxalNLP\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"google/WaxalNLP\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=google, license:cc-by-sa-4.0, size_categories:1M<n<10M, format:parquet, downloads=30916"],"format":"mixed","generatedAt":"2026-11-14T14:15:25.550Z","layout":{"example":"data/ASR/ach/ach-test-00000.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/ASR/ach/ach-test-00000.parquet","truncatedAt":512}],"sampleObjects":["croissant.json","data/ASR/ach/ach-test-00000.parquet","data/ASR/ach/ach-train-00000.parquet","data/ASR/ach/ach-train-00001.parquet","data/ASR/ach/ach-train-00002.parquet"],"sidecars":[{"bytes":"---\nlanguage_creators:\n  - creator_1\nlanguage:\n  - ach\n  - aka\n  - amh\n  # - bam\n  - bau\n  - dag\n  - dga\n  - ewe\n  - fat\n  # - fuf\n  - ful\n  - hau\n  - ibo\n  - kik\n  - kpo\n  - lin\n  - lug\n  - luo\n  - mas\n  - mlg\n  - nyn\n  - orm\n  - pcm\n  - sid\n  - sna\n  - sog\n  - swa\n  - tir\n  - twi\n  - wal\n  # - wol\n  - yor\nlicense:\n  - cc-by-sa-4.0\n  - cc-by-4.0\nmultilinguality:\n  - multilingual\nsource_datasets:\n  - UGSpeechData\n  - DigitalUmuganda/AfriVoice\n  - original\ntask_categories:\n  - automatic-speech-recognition\n  - text-to-speech\npretty_name: Waxal NLP Datasets\narxiv: 2602.02734\nannotation_creators:\n  - human-annotated\n  - crowdsourced\ntags:\n  - audio\n  - automatic-speech-recognition\n  - text-to-speech\nconfigs:\n  - config_name: ach_asr\n    data_files:\n      - split: train\n        path: data/ASR/ach/ach-train-*\n      - split: validation\n        path: data/ASR/ach/ach-validation-*\n      - split: test\n        path: data/ASR/ach/ach-test-*\n      - split: unlabeled\n        path: data/ASR/ach/ach-unlabeled-*\n  - config_name: ach_tts\n    data_files:\n      - split: train\n        path: data/TTS/ach/ach-train-*\n      - split: validation\n        path: data/TTS/ach/ach-validation-*\n      - split: test\n        path: data/TTS/ach/ach-test-*\n  - config_name: aka_asr\n    data_files:\n      - split: train\n        path: data/ASR/aka/aka-train-*\n      - split: validation\n        path: data/ASR/aka/aka-validation-*\n      - split: test\n        path: data/ASR/aka/aka-test-*\n      - split: unlabeled\n        path: data/ASR/aka/aka-unlabeled-*\n  - config_name: amh_asr\n    data_files:\n      - split: train\n        path: data/ASR/amh/amh-train-*\n      - split: validation\n        path: data/ASR/amh/amh-validation-*\n      - split: test\n        path: data/ASR/amh/amh-test-*\n      - split: unlabeled\n        path: data/ASR/amh/amh-unlabeled-*\n  # - config_name: bam_tts\n  #   data_files:\n  #     - split: train\n  #       path: data/TTS/bam/bam-train-*\n  #     - split: validation\n  #       path: data/TTS/bam/bam-validation-*\n  #     - split: test\n  #       path: data/TTS/bam/bam-test-*\n  - config_name: bau_tts\n    data_files:\n      - split: train\n        path: data/TTS/bau/bau-train-*\n      - split: validation\n        path: data/TTS/bau/bau-validation-*\n      - split: test\n        path: data/TTS/bau/bau-test-*\n  - config_name: dag_asr\n    data_files:\n      - split: train\n        path: data/ASR/dag/train-*\n      - split: validation\n        path: data/ASR/dag/validation-*\n      - split: test\n        path: data/ASR/dag/test-*\n      - split: unlabeled\n        path: data/ASR/dag/unlabeled-*\n  - config_name: dga_asr\n    data_files:\n      - split: train\n        path: data/ASR/dga/train-*\n      - split: validation\n        path: data/ASR/dga/validation-*\n      - split: test\n        path: data/ASR/dga/test-*\n      - split: unlabeled\n        path: data/ASR/dga/unlabeled-*\n  - config_name: ewe_asr\n    data_files:\n      - split: train\n        path: data/ASR/ewe/ewe-train-*\n      - split: validation\n        path: data/ASR/ewe/ewe-validation-*\n      - split: test\n        path: data/ASR/ewe/ewe-test-*\n      - split: unlabeled\n        path: data/ASR/ewe/ewe-unlabeled-*\n  - config_name: ewe_tts\n    data_files:\n      - split: train\n        path: data/TTS/ewe/ewe-train-*\n      - split: validation\n        path: data/TTS/ewe/ewe-validation-*\n      - split: test\n        path: data/TTS/ewe/ewe-test-*\n  - config_name: fat_tts\n    data_files:\n      - split: train\n        path: data/TTS/fat/fat-train-*\n      - split: validation\n        path: data/TTS/fat/fat-validation-*\n      - split: test\n        path: data/TTS/fat/fat-test-*\n  - config_name: ful_asr\n    data_files:\n      - split: train\n        path: data/ASR/ful/ful-train-*\n      - split: validation\n        path: data/ASR/ful/ful-validation-*\n      - split: test\n        path: data/ASR/ful/ful-test-*\n      - split: unlabeled\n        path: data/ASR/ful/ful-unlabeled-*\n  # - config_name: fuf_tts\n  #   data_files:\n  #     - split: train\n  #       path: data/TTS/fuf/fuf-train-*\n  #     - split: validation\n  #       path: data/TTS/fuf/fuf-vali","encoding":"utf8","key":"README.md","name":"README.md","size":36256,"truncated":true}],"sourceObjectKey":"data/ASR/ach/ach-test-00000.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"google/WaxalNLP\")\nprint(ds)","starterSamplePath":"data/ASR/ach/ach-test-00000.parquet","totalObjects":2619},"summary":"WAXAL is a large-scale multilingual speech corpus for African languages, released by Google with partner institutions and funded by Google and the Gates Foundation under an…","title":"WAXAL: multilingual African language speech corpus (ASR and TTS)","updatedAt":"2026-11-14T17:36:48.046Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/google/WaxalNLP"}}