datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
neural-mathrock
Neural Math Rock Multimodal Emotion Dataset
Dataset Description
This corpus is a large-scale multimodal emotion classification dataset specifically developed for Music Information Retrieval (MIR) and emotional computational analysis within complex musical genres, predominantly Math Rock and Midwest Emo. The dataset consists of exactly 4,000 distinct full-length tracks structured directly from the validated metadata registry.
The primary objective of this corpus is… See the full description on the dataset page: https://huggingface.co/datasets/anggars/neural-mathrock.VoiceAssistant-Eval
🔥 VoiceAssistant-Eval: Benchmarking AI Assistants across Listening, Speaking, and Viewing
[🌐 Homepage]
[🔮 Visualization]
[💻 Github]
[📖 Paper]
[📊 Leaderboard ]
[📊 Detailed Leaderboard ]
[📊 Roleplay Leaderboard ]
🚀 Data Usage
from datasets import load_dataset
for split in ['listening_general', 'listening_music', 'listening_sound', 'listening_speech',
'speaking_assistant', 'speaking_emotion', 'speaking_instruction_following'… See the full description on the dataset page: https://huggingface.co/datasets/MathLLMs/VoiceAssistant-Eval.MoulSot-Full
MoulSot-Full Dataset
Dataset Summary
MoulSot-Full is a large-scale Moroccan Darija speech dataset containing in total 1,500 hours of speech audio. From this extensive corpus, a high-quality subset of approximately 80 hours has been carefully curated and transcribed. It was built entirely from publicly available YouTube content across 51 diverse channels (including vlogs, podcasts, interviews, and commentary) to capture real-world Moroccan Darija, including natural… See the full description on the dataset page: https://huggingface.co/datasets/MathematicianNLPer/MoulSot-Full.khanacademy-turkish-mathMiniProjectMLThis dataset is designed for training an audio classification model that identifies the type of trash being thrown into a bucket.
The model classifies sounds into the following categories: Metal, Glass, Plastic, Cardboard, and Noise (non-trash-related sounds).
The dataset was recorded and organized as part of an Edge Impulse project to create a system that sorts trash based on sound.
Link to Edge Impulse: https://studio.edgeimpulse.com/public/556872/live
qaloon-reciter-dataset
Qaloon Reciter — continuously maintained data
Initial release: 0.1.0; schema 1.0.0. A private team working corpus for
Qālūn ʿan Nāfiʿ ASR. Current scope: al-Fātiḥah + Juzʾ ʿAmma (78–114), not
the whole Quran. Three reciters have been approved by the project owner for
this team release: Huthaify, Husary and Dokali. This approval is not a claim
of a verified source copyright license or acoustic alignment review.
Included and excluded material
569 recordings per… See the full description on the dataset page: https://huggingface.co/datasets/Mathani-Ayat/qaloon-reciter-dataset.french-math-asr-benchmarkMusicBoxmath-speech-datasetlumi_math_13-7huawei-mathMathSpeechqaloon-reciter-experiments
Experimental only — Waleed and Trabulsi
Private research inventory uploaded at the project owner's request. Not part
of the approved training dataset. Both sources retain authorization_status: unauthorized; an experimental upload is not source redistribution permission.
Waleed: 569 individually labelled WAVs, durations read from WAV headers;
Fātiḥah uses a separate basmalah track and fused ending. Source labels have
not been independently checked. Some source identity entries… See the full description on the dataset page: https://huggingface.co/datasets/Mathani-Ayat/qaloon-reciter-experiments.MathSpeech_whisper_transcribed_normalizedMathSpeech_whisper_transcribedHindi50_3math-audiomatheusaleixomathbridge-audio
Dataset Card for MathBridge_Audio
Dataset Description
This dataset is an expansion of the MathBridge dataset by Kyudan. The original dataset gives translations from English text to LaTeX formatting with context before and after. Our dataset is a subset of 1,000 rows from MathBridge with added audio files of speakers reading the full text.
Curated by: Abigail Pitcairn
Funded by: NSF and University of Southern Maine AIIR Lab
Language: English (en)
License: open-source… See the full description on the dataset page: https://huggingface.co/datasets/abby1492/mathbridge-audio.spokensquad_perturbacoesmathew1MathSpeechAudiosPortuguesMatheusasdMathSpeechspeech_mathHindi50Hindi50_2MathAllThis is the testmini set of MathAll.