Home › Datasets › modality › Speech
Speech datasets
archive 2025-07-28197 datasets carry the modality tag "Speech", ordered by the archive's paper count. Page 4 of 5: 48 shown of 197. Facet routes are this site's own (the archive records the tag string, not a page).
The archive holds 12,214 dataset rows; 12,172 are listed. 6 are withheld from every listing and count here as vandalised before snapshot (6 with contact-centre spam in the title, 0 with a spam description on a row that has no homepage, no paper and no papers counted; none with more than 1 paper, 0 with a benchmark), listed in withheld.json; 1 listed row carries a vandalised description, withheld on its page. This gate never withholds a row with a homepage or a paper that resolves, and a clean description; the content rules below withhold a row whose name is spam whatever else it carries. The gate is a phrase list: these are the rows it caught, not a claim that the rest is clean. Before that gate, the site's content rules withhold 36 more rows (invite-code, gambling, travel-booking, contact-centre and similar spam in the name or on a row with nothing real behind it); they have no page and are listed in withheld.json.
Filter 39 modality tags shown of 39, by dataset count; the full filter by modality, task and language is on /datasets
All 12,172
Images 3,239Texts 3,130Videos 1,014Audio 480Medical 3943D 380Time series 294Graphs 282Tabular 267Speech 197RGB-D 190Environment 147Point cloud 134Biomedical 122LiDAR 94RGB Video 87Tracking 77Biology 70Actions 673d meshes 64Tables 50Music 47EEG 44Hyperspectral images 44Stereo 44MRI 39Physics 32Interactive 29Dialog 25Midi 226D 17Replay data 11Financial 10Ranking 10Cad 9fMRI 7Parallel 6Lyrics 2PSG 2
Speech datasets 145–192 of 197
JSS Dataset (Jejueo Single Speaker Speech)
The Jejueo Single Speaker Speech (JSS) dataset consists of 10k high-quality audio files recorded by a native Jejueo speaker and a transcript file.
1 paper · 0 benchmarks
Kinect-WSJ is a multichannel, multispeaker, reverberated, noisy dataset which extends the WSJ0-2mix singlechannel, non-reverberated, noiseless dataset to the strong reverberation and noise conditions and the Kinect-like microphone array…
1 paper · 0 benchmarks
The Kite database is a multi-modal dataset for the control of unmanned aerial vehicles (UAVs).
1 paper · 0 benchmarks
LibriS2S is a Speech to Speech Translation (S2ST) dataset build further upon existing resources.
1 paper · 0 benchmarks
MC_GRID (Multi_Channel_Grid)
Here we release the dataset (MultiChannelGrid, abbreviated as MCGrid) used in our paper LIMUSE: LIGHTWEIGHT MULTI-MODAL SPEAKER EXTRACTION](https://arxiv.org/abs/2111.04063)).
1 paper · 0 benchmarks
MSNER (Multilingual Spoken Named Entity Recognition)
This dataset contains named entities annotations for European Parliament recordings in Dutch, French, German and Spanish.
1 paper · 0 benchmarks
MediBeng (Synthetic Code-Switched Bengali-English Speech Conversations for Healthcare Applications)
MediBeng Dataset The MediBeng dataset contains synthetic code-switched dialogues in Bengali and English for training models in speech recognition (ASR), text-to-speech (TTS), and machine translation in clinical settings.
1 paper · 1 benchmark
We announce the release of a new multilingual speaker dataset called NITK-IISc Multilingual Multi-accent Speaker Profiling(NISP) dataset.
1 paper · 0 benchmarks
The NISQA Corpus includes more than 14,000 speech samples with simulated (e.g.
1 paper · 0 benchmarks
Parkinson Speech Dataset is an audio dataset consisting of recordings of 20 Parkinson's Disease (PD) patients and 20 healthy subjects.
1 paper · 0 benchmarks
Data collection was conducted by asking some adults from social media and some students from an elementary school to participate in our experiment.
1 paper · 0 benchmarks
Quechua Collao corpus for automatic emotion recognition in speech.
1 paper · 1 benchmark
Speech Recognition Dataset for Oromo Language.
1 paper · 1 benchmark
Facial electromyography recordings during both silent and vocalized speech.
1 paper · 0 benchmarks
Dataset Summary Speech Brown is a comprehensive, synthetic, and diverse paired speech-text dataset in 15 categories, covering a wide range of topics from fiction to religion.
1 paper · 0 benchmarks
Spot the Difference Corpus is a corpus of task-oriented spontaneous dialogues which contains 54 interactions between pairs of subjects interacting to find differences in two very similar scenes.
1 paper · 0 benchmarks
TAT (Taiwanese Across Taiwan)
Taiwanese Across Taiwan (TAT) corpus is a Large-Scale database of Native Taiwanese Article/Reading Speech collected across Taiwan.
1 paper · 1 benchmark
The TED VCR Video Retrieval Dataset is a multimodal collection derived from publicly available TED Talks.
1 paper · 0 benchmarks
Dubbed series are gaining a lot of popularity in recent years with strong support from major media service providers.
1 paper · 0 benchmarks
100 samples each of synthetic speech generated by 9 moderns TTS systems.
1 paper · 0 benchmarks
The SWC is a corpus of aligned Spoken Wikipedia articles from the English, German, and Dutch Wikipedia.
1 paper · 1 benchmark
The United-Syn-Med dataset is a specialized medical speech dataset designed to evaluate and improve Automatic Speech Recognition (ASR) systems within the healthcare domain.
1 paper · 0 benchmarks
Uses same clean speech as VoiceBank+Demand but more noise types.
1 paper · 1 benchmark
VESUS (Varied Emotion in Syntactically Uniform Speech)
The Varied Emotion in Syntactically Uniform Speech (VESUS) repository is a lexically controlled database collected by the NSA lab.
1 paper · 0 benchmarks
VedantaNY-10M is a curated dataset of over 750 hours of transcripts from public discourses on the Indian philosophy of Advaita Vedanta.
1 paper · 0 benchmarks
This is the forehead accelerometer variant of the VibraVox dataset.
1 paper · 3 benchmarks
This is the reference headset microphone variant of the VibraVox dataset.
1 paper · 2 benchmarks
This is the in-ear rigid earpiece-embedded microphone variant of the VibraVox dataset.
1 paper · 3 benchmarks
This is the in-ear comply foam-embedded microphone variant of the VibraVox dataset.
1 paper · 3 benchmarks
This is the temple vibration pickup variant of the VibraVox dataset.
1 paper · 3 benchmarks
This is the throat microphone (laryngophone) variant of the VibraVox dataset.
1 paper · 3 benchmarks
Voice Navigation is a large-scale dataset of Chinese speech for slot filling, containing more than 830,000 samples.
1 paper · 0 benchmarks
1 paper · 1 benchmark
Vāksañcayaḥ (Sanskrit Speech Corpus by IIT Bombay)
This Sanskrit speech corpus has more than 78 hours of audio data and contains recordings of 45,953 sentences with a sampling rate of 22KHz.
1 paper · 0 benchmarks
WHAMRext is an extension to the WHAMR corpus with larger RT60 values (between 1s and 3s)
1 paper · 1 benchmark
The Watch Your Mouth dataset is a custom silent speech dataset consisting of depth-only recordings of users silently mouthing full English sentences, captured using consumer-grade depth cameras such as the iPhone TrueDepth sensor.
1 paper · 0 benchmarks
Well-being Dataset (Cambridge Well-being Dataset for Psychological Distress Analysis)
The dataset is a private dataset collected for automatic analysis of psychological distress.
1 paper · 1 benchmark
inaGVAD (InaGVAD : a Challenging French TV and Radio Corpus annotated for Voice Activity Detection and Speaker Gender Segmentation)
InaGVAD is a Voice Activity Detection (VAD) and Speaker Gender Segmentation (SGS) dataset designed for representing the acoustic diversity of French TV and Radio programs.
1 paper · 0 benchmarks
A modification on the ShEMO dataset with help of an Automatic Speech Recognition (ASR) system.
1 paper · 0 benchmarks
1 paper · 0 benchmarks
The Arabic Speech Corpus (1.5 GB) is a Modern Standard Arabic (MSA) speech corpus for speech synthesis.
0 papers · 0 benchmarks
BABEL is a multilingual corpus of conversational telephone speech from IARPA, which includes Asian and African languages.
0 papers · 0 benchmarks
The CMU Wilderness Multilingual Speech Dataset is a dataset of over 700 different languages providing audio, aligned text and word pronunciations.
0 papers · 0 benchmarks
Deeply Korean read speech corpus contains pairs of Korean speakers reading a script with 3 distinct text sentiments (negative, neutral, positive), with 3 distinct voice sentiments (negative, neutral, positive), are recorded.
0 papers · 0 benchmarks
Deeply Parent-Child Vocal Interaction contains the interaction of 24 pairs of parent and child(total 48 speakers), such as reading fairy tales, singing children’s songs, conversing, and others, is recorded.
0 papers · 0 benchmarks
Deeply vocal characterizer is a human nonverbal vocalization dataset.
0 papers · 0 benchmarks
0 papers · 0 benchmarks
JTES (Japanese Twitter-based Emotional Speech)
We designed an emotional speech database that can be used for emotion recognition as well as recognition and synthesis of speech with various emotions.
0 papers · 0 benchmarks
Paper counts and descriptions are the archive's, frozen 2025-07-28; no citation counts, no stars, no trending. Sorting by "most cited" or "newest" was a live-site feature the archive does not carry.