datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Multitask-National-Speech-Corpus-v1Multitask-National-Speech-Corpus (MNSC v1) is derived from IMDA's NSC Corpus.
MNSC is a multitask speech understanding dataset derived and further annotated from IMDA NSC Corpus. It focuses on the knowledge of Singapore's local accent, localised terms, and code-switching.
ASR: Automatic Speech Recognition
SQA: Speech Question Answering
SDS: Spoken Dialogue Summarization
PQA: Paralinguistic Question Answering
from datasets import load_dataset
data =… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/Multitask-National-Speech-Corpus-v1.sea_audiobench_datasets_ASR
SEA-SpeechBench — ASR (Automatic Speech Recognition)
This dataset is the automatic-speech-recognition (ASR) task of
SEA-SpeechBench, a large-scale multitask benchmark for speech
understanding across Southeast Asia. It contains 26,863 evaluation examples
across eleven languages, drawn from fifteen source corpora, each pairing an
audio recording with an instruction and a reference transcript.
Given the recording and the instruction, a model must transcribe the speech.… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_ASR.sea_audiobench_datasets_TCQ
SEA-SpeechBench — TCQ (Timestamped Content Query)
This dataset is the timestamped-content-query (TCQ) task of
SEA-SpeechBench, a large-scale multitask benchmark for speech
understanding across Southeast Asia. It contains 14,172 evaluation examples
across five languages, each pairing an audio recording with an instruction
and a reference answer.
Given the recording and a timestamp, a model must report what is said at
that point in the audio. Contexts run from 30 seconds to 3… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_TCQ.sea_audiobench_datasets_PQA
SEA-SpeechBench — Paralinguistics (AGE, ER, GR)
This dataset holds the three paralinguistic tasks of SEA-SpeechBench, a
large-scale multitask benchmark for speech understanding across Southeast
Asia. It contains 25,563 evaluation examples across nine languages, drawn from
fourteen source corpora, each pairing an audio recording with an instruction
and a reference answer.
Given the recording and the instruction, a model must identify a property of
the speaker or the delivery… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_PQA.sea_audiobench_datasets_TLoc
SEA-SpeechBench — TLoc (Temporal Localization)
This dataset is the temporal-localization (TLoc) task of
SEA-SpeechBench, a large-scale multitask benchmark for speech
understanding across Southeast Asia. It contains 6,736 evaluation examples
across five languages, each pairing an audio recording of 30–180 seconds
with an instruction and a reference answer.
Given the recording and the instruction, a model must identify when a
described utterance occurs.
Quick start… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_TLoc.sea_audiobench_datasets_SQA
SEA-SpeechBench — SQA (Spoken Question Answering)
This dataset is the spoken-question-answering (SQA) task of
SEA-SpeechBench, a large-scale multitask benchmark for speech
understanding across Southeast Asia. It contains 5,462 evaluation examples
across five languages, each pairing an audio recording with a question and a
reference answer.
Given the recording and the question, a model must answer using the content
of the speech.
Quick start
Requires datasets>=4.0… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_SQA.sea_audiobench_datasets_SpR
SEA-SpeechBench — SpkR (Speaker Recognition)
This dataset is the speaker-recognition (SpkR) task of SEA-SpeechBench,
a large-scale multitask benchmark for speech understanding across Southeast
Asia. It contains 8,827 evaluation examples across eight languages, each
pairing two audio recordings with an instruction and a reference answer.
Given the two recordings and the instruction, a model must compare the
speakers in them.
Quick start
Requires datasets>=4.0… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_SpR.sea_audiobench_datasets_ST
SEA-SpeechBench — ST (Speech Translation)
This dataset is the speech-translation (ST) task of SEA-SpeechBench, a
large-scale multitask benchmark for speech understanding across Southeast
Asia. It contains 7,189 evaluation examples across nine source languages,
each pairing an audio recording with an instruction and a reference
translation.
Given the recording and the instruction, a model must translate the speech
into the target language.
Quick start
Requires… See the full description on the dataset page: https://huggingface.co/datasets/MERaLiON/sea_audiobench_datasets_ST.