datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
biggest-ru-bookA bigger version of its5Q/bigger-ru-book, the smaller set being a subset of this one. Almost 1000 hours of high-quality audio.
sepedinova-rvc-datasetbigger-ru-bookamharic-speech-dataset-2026
Amharic Speech Dataset 2026
Overview
This dataset contains Amharic speech recordings collected using the Leyu Platform for the Leyu Platform Competition 2026.
Language
Amharic (am)
Dialect
Standard Addis Ababa Amharic
Speaker Information
Number of Speakers: 1
Speaker IDs: SPK001
Audio Format
Format: M4A
Duration: 10–60 seconds per recording
Directory Structure
audio/
metadata.csv… See the full description on the dataset page: https://huggingface.co/datasets/ofc-its-phyla/amharic-speech-dataset-2026.english-asr-processedHTMOneShotLoopClassification
Dataset Card for HTMOneShotLoopClassification
Dataset Summary
HTMOneShotLoopClassification is a dataset of 5,561 electronic music samples (one-shots and loops) from house, tech house, and minimal techno genres. Each sample is labeled with one of 8 stem categories and includes 55 extracted audio features for machine learning applications.
Supported Tasks and Leaderboards
Audio Classification: Multi-class classification into 8 stem categories
Baseline… See the full description on the dataset page: https://huggingface.co/datasets/itsuzef/HTMOneShotLoopClassification.hoping_its_final_datasetddaa-sample
DDAA Sample Release
This is a representative sample of the full DDAA release for reviewer inspection.
Contents:
unified_dataset split manifests and metadata-aligned CSVs
one representative official_dataset processed shard
Sampling method:
Includes one processed audio shard and canonical split manifests (train.csv, val.csv, test.csv)
Preserves directory layout and schema used by the full dataset.
itsuvoiceindian-tts-dataset
Indian TTS Dataset — Hindi + Indian English
A curated, single-speaker TTS training dataset with 144 segments (~60 minutes total) sourced from YouTube, transcribed using Sarvam AI ASR, and annotated with emotion/style tags.
Dataset Summary
Split
Segments
Duration
Indian English
71
29.6 min
Hindi
73
30.8 min
Total
144
60.3 min
Audio Specs
Sample rate: 16 kHz
Channels: Mono
Format: WAV (16-bit PCM)
Segment length: 20–28 seconds… See the full description on the dataset page: https://huggingface.co/datasets/Itsharshi/indian-tts-dataset.KTT-Day3
🧠 Edge-Optimized Math Tutor Orchestrator
Model Description
This repository contains the orchestration and adaptive logic layer for an offline, CPU-bound AI Math Tutor. Rather than acting as a standalone monolithic weight file, this pipeline links extreme-edge open-source models with Bayesian logic to meet a strict < 75MB operating footprint constraint.
Core Architecture
ASR Component: openai/whisper-tiny (39M parameters). Handles primary transcription and… See the full description on the dataset page: https://huggingface.co/datasets/itsazza/KTT-Day3.tunision_number_banking_setRVmodelRmodelrootdatasetearnings21-pause-smalliajesusaudiomifos_audio_benchmarking