Audio & Speech datasets

Audio and speech recordings for speech, speaker and music recognition.

25 results

Download the results' metadata: CSV JSON
Audio & Speech Partially verified

AccentDB

AccentDB is a multi-pairwise parallel corpus of structured and labelled accented speech. It contains speech samples from speakers of 4 non-native accents of English (8 speakers,…

Samples
17,313
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Needs review

CMU ARCTIC

Speech-synthesis databases of single-speaker English recordings (18 speaker sets available in torchaudio).

Samples
Not available
Type
Audio
Licence
Unknown
Source
Publisher website
Audio & Speech Partially verified

Common Voice

Mozilla Common Voice Dataset

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

CREMA-D

CREMA-D is an audio-visual data set for emotion recognition. The data set consists of facial and vocal emotional expressions in sentences spoken in a range of…

Samples
7,438
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Needs review

Device Recorded VCTK (DR-VCTK)

Device Recorded VCTK (small subset version): VCTK speech re-recorded through consumer devices, with train and test subsets.

Samples
Not available
Type
Audio
Licence
Unknown
Source
Publisher website
Audio & Speech Partially verified

Free Spoken Digit Dataset

A free audio dataset of spoken digits. Think MNIST for audio.

Samples
2,500
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

FUSS (Free Universal Sound Separation)

The Free Universal Sound Separation (FUSS) Dataset is a database of arbitrary sound mixtures and source-level references, for use in experiments on arbitrary sound separation. This…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

Groove MIDI Dataset

The Groove MIDI Dataset (GMD) is composed of 13.6 hours of aligned MIDI and (synthesized) audio of human-performed, tempo-aligned expressive drumming captured on a Roland TD-11…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

GTZAN Genre Collection

The dataset consists of 1000 audio tracks each 30 seconds long. It contains 10 genres, each represented by 100 tracks.

Samples
1,000
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

GTZAN Music/Speech

The dataset was collected for the purposes of music/speech discrimination. The dataset consists of 120 tracks, each 30 seconds long.

Samples
128
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Needs review

Libri-Light (limited supervision subset)

Subset of Libri-light used in HuBERT for supervised fine-tuning, with 10-minute, 1-hour and 10-hour labelled subsets.

Samples
Not available
Type
Audio
Licence
Unknown
Source
Publisher website
Audio & Speech Needs review

LibriMix

Speech separation dataset with Libri2Mix and Libri3Mix mixtures (train-360, train-100, dev, test).

Samples
Not available
Type
Audio
Licence
Unknown
Source
Publisher website
Audio & Speech Partially verified

LibriSpeech

LibriSpeech is a corpus of approximately 1000 hours of read English speech with sampling rate of 16 kHz, prepared by Vassil Panayotov with the assistance of…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

LibriTTS

LibriTTS is a multi-speaker English corpus of approximately 585 hours of read English speech at 24kHz sampling rate, prepared by Heiga Zen with the assistance of…

Samples
375,086
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

LJSpeech

This is a public domain speech dataset consisting of 13,100 short audio clips of a single speaker reading passages from 7 non-fiction books. A transcription is…

Samples
13,100
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Needs review

MUSDB18-HQ

Uncompressed version of MUSDB18 for music source separation, with bass, drums, other, vocals and mixture stems.

Samples
Not available
Type
Audio
Licence
Unknown
Source
Zenodo
Audio & Speech Partially verified

NSynth

The NSynth Dataset is an audio dataset containing ~300k musical notes, each with a unique pitch, timbre, and envelope. Each note is annotated with three additional…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Needs review

QUESST14

Query-by-example spoken term detection database in several languages (albanian, basque, czech, nnenglish, romanian, slovak).

Samples
Not available
Type
Audio
Licence
Unknown
Source
Publisher website
Audio & Speech Partially verified

SAVEE

SAVEE (Surrey Audio-Visual Expressed Emotion) is an emotion recognition dataset. It consists of recordings from 4 male actors in 7 different emotions, 480 British English utterances…

Samples
480
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

Speech Commands

An audio dataset of spoken words designed to help train and evaluate keyword spotting systems. Its primary goal is to provide a way to build and…

Samples
100,503
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

TED-LIUM

The TED-LIUM corpus is English-language TED talks, with transcriptions, sampled at 16kHz. It contains about 118 hours of speech.

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

VCTK

This CSTR VCTK Corpus includes speech data uttered by 110 English speakers with various accents. Each speaker reads out about 400 sentences, which were selected from…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

VoxCeleb

An large scale dataset for speaker identification. This data is collected from over 1,251 speakers, with over 150k samples in total.

Samples
148,642
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog
Audio & Speech Partially verified

XTREME-S

FLEURS is the speech version of the FLORES machine translation benchmark, covering 2000 n-way parallel sentences in n=102 languages. XTREME-S covers four task families: speech recognition,…

Samples
Not available
Type
Audio
Licence
Unknown
Source
TensorFlow Datasets catalog