{"success":true,"data":{"author":"Evelina Bakhturina, Vitaly Lavrukhin, Boris Ginsburg, Yang Zhang (NVIDIA)","bucketAccessType":"public","bucketName":"MikhailT/hifi-tts","bucketPrefix":"","category":"Speech & Audio","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:17.397Z","description":"Hi-Fi TTS is a multi-speaker English dataset for training text-to-speech models, built by NVIDIA from LibriVox public domain audiobooks and Project Gutenberg texts. It contains about 291.6 hours of speech from 10 speakers, each with at least 17 hours, sampled at 44.1 kHz, which makes it one of the few open corpora with enough high-sample-rate audio per speaker to train studio-quality single- and multi-speaker TTS and neural vocoders. The original release sorts each speaker's audio into 'clean' and 'other' quality bins. This repository is a Parquet conversion of the OpenSLR SLR109 release: a data folder of 93 Parquet shards named by split and quality (for example dev.clean, dev.other, test.clean, train.clean), each row holding the audio and its text (measured 2026-11-14; the Hub reports about 166 GB). Open with datasets.load_dataset('MikhailT/hifi-tts') or read the shards with pyarrow or pandas. The dataset is released by its authors on OpenSLR under CC BY 4.0; cite Bakhturina et al., 'Hi-Fi Multi-Speaker English TTS Dataset' (arXiv 2104.01497).","displayTitle":"Hi-Fi Multi-Speaker English TTS Dataset (Hi-Fi TTS)","id":"ab3f3976-1de0-50db-a417-308c6a30e2d9","license":"CC-BY-4.0","licenseEvidenceUrl":"https://www.openslr.org/109/","objectCountEstimate":95,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2104.01497","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"MikhailT/hifi-tts\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"MikhailT/hifi-tts\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=MikhailT, license:cc-by-4.0, size_categories:100K<n<1M, format:parquet, downloads=4384"],"format":"mixed","generatedAt":"2026-11-14T14:15:19.520Z","layout":{"example":"data/dev.clean-00000-of-00001-2202c1655cccc3db.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"504152311504152015244c15041500120000103c0400000036303937040000003930313715001518151c2c15c8011510150615061c36002804393031371804363039370000000c2c03000000c80101016400640126a8011c150c1935100006191807737065616b6572150216c80116980116a001264826081c360028043930313718043630393700192c15041500150200150015101502000000150415965c15b4174c15c80115001200008b2ee035000000617564696f2f363039375f636c65616e2f31343431312f6e6164615f6c696c795f30305f686167676172645f303437302e666c6163ca390004312ea639000038197208333238aa39000039193908323237d63900043334aae400043136197208313034a6720004333319390835373205390034528f016c353332362f686172746d616e6e5f30325f666177636574745f3033362955be3800043430aa38000035197004323709e19270000036193808303636a6e0002e38002d1800315618015c3237372f737761675f30335f746f6d706b696e735f30323029f8b63500003829f482350000341d6a00354d9b82350000361d35043137290c003a52d40088313237312f676c6974746572696e67706c61696e5f31375f6d6f727269735f30313835254a0030523e0020333635372f726f6f74012a1534083235339a3400003215340030a26800199c00324d8e829c","encoding":"hex","key":"data/dev.clean-00000-of-00001-2202c1655cccc3db.parquet","truncatedAt":512}],"sampleObjects":["data/dev.clean-00000-of-00001-2202c1655cccc3db.parquet","data/dev.other-00000-of-00001-331a8f0a0261df76.parquet","data/test.clean-00000-of-00001-ad9def4d041642e2.parquet","data/test.other-00000-of-00001-1e57e6919880d8dd.parquet","data/train.clean-00000-of-00035-9a6c3c024271def3.parquet"],"sidecars":[{"bytes":"---\nconfigs:\n- config_name: clean\n  version: 1.0.0\n  data_files:\n  - split: train\n    path: data/train.clean-*\n  - split: test\n    path: data/test.clean-*\n  - split: dev\n    path: data/dev.clean-*\n- config_name: other\n  version: 1.0.0\n  data_files:\n  - split: train\n    path: data/train.other-*\n  - split: test\n    path: data/test.other-*\n  - split: dev\n    path: data/dev.other-*\n- config_name: all\n  version: 1.0.0\n  data_files:\n  - split: train.clean\n    path: data/train.clean-*\n  - split: train.other\n    path: data/train.other-*\n  - split: dev.clean\n    path: data/dev.clean-*\n  - split: dev.other\n    path: data/dev.other-*\n  - split: test.clean\n    path: data/test.clean-*\n  - split: test.other\n    path: data/test.other-*\ndataset_info:\n- config_name: clean\n  features:\n  - name: speaker\n    dtype: string\n  - name: file\n    dtype: string\n  - name: duration\n    dtype: float32\n  - name: text\n    dtype: string\n  - name: text_no_preprocessing\n    dtype: string\n  - name: text_normalized\n    dtype: string\n  - name: audio\n    dtype:\n      audio:\n        sampling_rate: 44100\n  splits:\n  - name: train\n    num_bytes: 17023899243\n    num_examples: 125989\n  - name: dev\n    num_bytes: 24204633\n    num_examples: 150\n  - name: test\n    num_bytes: 52040552\n    num_examples: 300\n  download_size: 16271001158\n  dataset_size: 17104553676\n- config_name: other\n  features:\n  - name: speaker\n    dtype: string\n  - name: file\n    dtype: string\n  - name: duration\n    dtype: float32\n  - name: text\n    dtype: string\n  - name: text_no_preprocessing\n    dtype: string\n  - name: text_normalized\n    dtype: string\n  - name: audio\n    dtype:\n      audio:\n        sampling_rate: 44100\n  splits:\n  - name: train\n    num_bytes: 26755286687\n    num_examples: 196489\n  - name: dev\n    num_bytes: 65601521\n    num_examples: 350\n  - name: test\n    num_bytes: 129348882\n    num_examples: 700\n  download_size: 25655017468\n  dataset_size: 26957939607\n- config_name: all\n  features:\n  - name: speaker\n    dtype: string\n  - name: file\n    dtype: string\n  - name: duration\n    dtype: float32\n  - name: text\n    dtype: string\n  - name: text_no_preprocessing\n    dtype: string\n  - name: text_normalized\n    dtype: string\n  - name: audio\n    dtype:\n      audio:\n        sampling_rate: 44100\n  splits:\n  - name: train.clean\n    num_bytes: 17023899243\n    num_examples: 125989\n  - name: train.other\n    num_bytes: 26755286687\n    num_examples: 196489\n  - name: dev.clean\n    num_bytes: 24204633\n    num_examples: 150\n  - name: dev.other\n    num_bytes: 65601521\n    num_examples: 350\n  - name: test.clean\n    num_bytes: 52040552\n    num_examples: 300\n  - name: test.other\n    num_bytes: 129348882\n    num_examples: 700\n  download_size: 7040649041\n  dataset_size: 44050381518\npretty_name: HiFi TTS\ndescription: >-\n  Hi-Fi Multi-Speaker English TTS Dataset (Hi-Fi TTS) is based on LibriVox's\n  public domain audio books and Gutenberg Project texts.\nhomepage: http://www.openslr.org/109\nlanguage:\n- en\nlicense:\n- cc-by-4.0\ncitation: |\n  @article{bakhturina2021hi,\n    title={{Hi-Fi Multi-Speaker English TTS Dataset}},\n    author={Bakhturina, Evelina and Lavrukhin, Vitaly and Ginsburg, Boris and Zhang, Yang},\n    journal={arXiv preprint arXiv:2104.01497},\n    year={2021}\n  }\ntask_categories:\n- text-to-speech\n- text-to-audio\n---\n\n# Dataset Card for HiFiTTS\n\nHi-Fi Multi-Speaker English TTS Dataset (Hi-Fi TTS) is based on LibriVox's public domain audio books and Gutenberg Project texts.","encoding":"utf8","key":"README.md","name":"README.md","size":3457,"truncated":false}],"sourceObjectKey":"data/dev.clean-00000-of-00001-2202c1655cccc3db.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"MikhailT/hifi-tts\")\nprint(ds)","starterSamplePath":"data/dev.clean-00000-of-00001-2202c1655cccc3db.parquet","totalObjects":95},"summary":"Hi-Fi TTS is a multi-speaker English dataset for training text-to-speech models, built by NVIDIA from LibriVox public domain audiobooks and Project Gutenberg texts.","title":"Hi-Fi Multi-Speaker English TTS Dataset (Hi-Fi TTS)","updatedAt":"2026-11-14T17:33:52.363Z","visibility":"public","websiteUrl":"https://www.openslr.org/109/"}}