datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
danish-asr-unified-hviske-v5-tiny
danish-asr-unified — two-model labels and a quality manifest
Transcriptions, per-token confidences, and a per-row quality verdict for every
row of syvai/danish-asr-unified
(3,414,589 rows, 8 sources).
Two independently trained models labelled the whole corpus:
model
architecture
vocabulary
syvai/hviske-v5-tiny
encoder-decoder
16,384 BPE
3dio-ai/svale-110M
RNN-T (Parakeet)
44 characters
Both decode greedily. Shards mirror the source data/train-*.parquet by name… See the full description on the dataset page: https://huggingface.co/datasets/syvai/danish-asr-unified-hviske-v5-tiny.tr-subset-v0.1
TR Subset v0.1 — Turkish speech
251,118 Turkish audio/text rows (~62 GB, 128 parquet shards). Schema is just
text + audio; see the YAML header above.
An early-phase Turkish speech collection from the TinyAya data pipeline. It is
not part of the v0.3 Stage-2 training corpus — that is
tr-hi-mimi-encoded.
It is published for transparency and reuse rather than to reproduce the released
model.
from datasets import load_dataset
ds = load_dataset("tiny-aya-translate/tr-subset-v0.1"… See the full description on the dataset page: https://huggingface.co/datasets/tiny-aya-translate/tr-subset-v0.1.hinglish-casual
Hinglish Casual Speech
33,275 casual Hindi-English code-switched utterances (~31 GB) with audio,
transcripts in both Devanagari and Latin script (utterance /
utterance_latin), speaker ids, style metadata and durations. Full schema is in
the YAML header above.
Collected during the TinyAya programme to probe code-switched speech, which
neither the FLORES-derived text nor the TTS corpora cover. It is not part of
the v0.3 Stage-2 training set — that is
tr-hi-mimi-encoded.
from… See the full description on the dataset page: https://huggingface.co/datasets/tiny-aya-translate/hinglish-casual.fleurs-tr-hi-parallel-speech
FLEURS TR↔HI Parallel Speech
Turkish⇄Hindi parallel speech built from FLEURS — the real human speech
counterpart to this project's synthetic TTS corpora.
audio/
~8,935 clips
fleurs/
2,440 source FLEURS files
manifests/
selection + QC manifests (incl. accepted.jsonl)
Mimi-encoded downstream as
fleurs-tr-hi-mimi-encoded,
which is what the v0.3 evaluation actually consumed.
⚠️ Acoustic shift, not held-out text
An overlap audit of the derived… See the full description on the dataset page: https://huggingface.co/datasets/tiny-aya-translate/fleurs-tr-hi-parallel-speech.cv-tr-eval
Common Voice Turkish Eval
4,825 Turkish test clips (~45 MB, 16 kHz) in the Mozilla Common Voice schema:
transcription, duration, up_votes / down_votes, and the age / gender
/ accent speaker attributes. Schema in the YAML header above.
An evaluation-only Turkish counterpart to
lahaja-eval;
never trained on. Used to sanity-check Turkish ASR quality on real human
speech, which matters here because the v0.3 training corpus is entirely
synthetic TTS and the released model is… See the full description on the dataset page: https://huggingface.co/datasets/tiny-aya-translate/cv-tr-eval.CAESAR-TINY
Dataset Card for CAESAR-TINY
Dataset Summary
CAESAR-TINY is a synthetic code-switched dataset generated by combining monolingual samples in Catalan and Spanish.
The process includes trimming silences, normalizing audio volume, and introducing random pauses. It contains 2 hours of speech data, created by concatenating audio from the Common voice 17 Benchmark split and VoxForge Spanish datasets.
Example Usage
To load CAESAR-TINY:
from datasets import… See the full description on the dataset page: https://huggingface.co/datasets/BSC-LT/CAESAR-TINY.lahaja-eval
LAHAJA Hindi ASR Eval
3,076 Hindi test utterances (~712 MB) carrying rich speaker metadata —
native_language, native_state, gender, age_group, scenario — plus both
verbatim and normalized transcripts. Schema in the YAML header above.
Held as an evaluation set only: never trained on. Its dialect and
native-state labels make it useful for checking whether Hindi ASR quality holds
across accents rather than only on the average.
This is the benchmark behind hindi-tts-probe, which… See the full description on the dataset page: https://huggingface.co/datasets/tiny-aya-translate/lahaja-eval.