datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ghana-speech-ipa
Ghana Speech — Audio with IPA Transcripts
Speech with both transcript forms: the original orthography and the IPA phoneme
sequence read off the audio by ASR. Each language is a subset, with real
train/validation splits.
from datasets import load_dataset
ds = load_dataset("ghanaopendata/ghana-speech-ipa", "Akuapem_Twi_twi", split="train")
ds[0]["audio"] # decoded waveform, 16 kHz
ds[0]["text"] # original orthography
ds[0]["ipa"] # IPA phonemes
369,347 clips · ~747 h… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/ghana-speech-ipa.new-twi-tts-aligned-ipa
new-twi-tts-aligned + IPA phonemes
ghanaopendata/new-twi-tts-aligned with a machine-generated IPA phoneme
transcription for every clip, produced with
ghananlpcommunity/ghana-speech-phoneme-asr.
Audio included — this is self-contained, no join with the source dataset needed.
Contents
split
clips
hours
phoneme units
mean units/clip
test
16,140
17.24
663,140
41.1
train
145,258
155.21
5,945,389
40.9
Columns
column
type
meaning… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/new-twi-tts-aligned-ipa.ghana-speech-ipa
Ghana Speech — Audio with IPA Transcripts
Speech with both transcript forms: the original orthography and the IPA phoneme
sequence read off the audio by ASR. Each language is a subset, with real
train/validation splits.
from datasets import load_dataset
ds = load_dataset("ghanaopendata/ghana-speech-ipa", "Akuapem_Twi_twi", split="train")
ds[0]["audio"] # decoded waveform, 16 kHz
ds[0]["text"] # original orthography
ds[0]["ipa"] # IPA phonemes
369,347 clips · ~747 h… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/ghana-speech-ipa.ghana-english-speech-ipa
Ghanaian English Speech — Audio with IPA Transcripts
Speech with both transcript forms: the original orthography and the IPA phoneme
sequence read off the audio by ASR. Each language is a subset, with real
train/validation splits.
from datasets import load_dataset
ds = load_dataset("ghanaopendata/ghana-english-speech-ipa", "English_eng", split="train")
ds[0]["audio"] # decoded waveform, 16 kHz
ds[0]["text"] # original orthography
ds[0]["ipa"] # IPA phonemes
52,855… See the full description on the dataset page: https://huggingface.co/datasets/ghanaopenai/ghana-english-speech-ipa.wiktionary-ipa-audio-en
English Wiktionary IPA + audio
English pronunciation rows extracted from the structured Kaikki/Wiktextract
English dump, restricted to English entries with both IPA and a playable
Wikimedia Commons recording. The dataset contains one row per pronunciation
and recording pairing; an audio recording can therefore occur in more than
one row when Wiktionary associates it with multiple IPA or entry records.
Fields
The audio column is created by the Hugging Face… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/wiktionary-ipa-audio-en.african-speech-ipa
African Speech IPA
AfriSpeech audio
paired with IPA phoneme transcriptions, for 141 languages.
Each row has the audio, the source transcript, and ipa — the transcript converted to
space-separated IPA phoneme units with
africa-g2p 0.2.0. Units are kept whole, so
t͡ʃ, k͡p and kʰ are one token each rather than two or three characters, and
punctuation is preserved as its own unit so alignment and TTS keep their phrasing.
from datasets import load_dataset
ds =… See the full description on the dataset page: https://huggingface.co/datasets/AfriSpeech/african-speech-ipa.new-twi-tts-aligned-ipa
new-twi-tts-aligned + IPA phonemes
ghanaopendata/new-twi-tts-aligned with a machine-generated IPA phoneme
transcription for every clip, produced with
ghananlpcommunity/ghana-speech-phoneme-asr.
Audio included — this is self-contained, no join with the source dataset needed.
Contents
split
clips
hours
phoneme units
mean units/clip
test
16,140
17.24
663,140
41.1
train
145,258
155.21
5,945,389
40.9
Columns
column
type
meaning… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/new-twi-tts-aligned-ipa.twi-health-asr-gemini-500hrs-ipa
Twi Health Speech — Audio, Transcript and IPA
Twi health-domain speech with both a written transcript and an IPA phoneme sequence read off the audio by ASR. Built from ghananlpcommunity/twi-health-asr-gemini-500hrs by adding the IPA column.
from datasets import load_dataset
ds = load_dataset("ghananlpcommunity/twi-health-asr-gemini-500hrs-ipa", split="train")
ds[0]["audio"] # decoded waveform, 16 kHz
ds[0]["transcription"] # transcript
ds[0]["ipa"]… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/twi-health-asr-gemini-500hrs-ipa.barranquenho-ipa-dict-synthetic
Barranquenho IPA Pronunciation Dictionary
The first and only IPA pronunciation dictionary of Barranquenho — the
Ibero-Romance contact variety spoken in Barrancos (Baixo Alentejo, Portugal),
a mixed system born of centuries of Portuguese–Spanish (Extremaduran /
Andalusian) contact on the raia. Every headword is written in the
Convenção Ortográfica do Barranquenho (2025) orthography and paired with a
broad-phonemic IPA transcription plus Portuguese and Spanish glosses.
This… See the full description on the dataset page: https://huggingface.co/datasets/TigreGotico/barranquenho-ipa-dict-synthetic.ipapack_plus_clean
IPAPACK++ Cleaned-Label Overlay (v1)
A labels-only, strictly additive cleaned overlay on IPAPACK++ (Zhu et al., ZIPA, ACL 2025). It adds a regenerated phoneme label (phones_v1_clean) alongside the paper's original phones, a drop_flag for the rows we could not repair, a per-utterance sigs list of which defect (if any) was fixed, and a per-row license tag. The original paper label is preserved on every row, including the dropped ones — so you can compare, ablate, or refuse the… See the full description on the dataset page: https://huggingface.co/datasets/sejongwang/ipapack_plus_clean.german-golden-audio_speech-IPA
🌟 German Golden Speech & IPA Corpus (FLEURS + Multilingual TEDx)
An ultra-clean, high-standard curated German speech dataset combining Google FLEURS (de_de) and Multilingual TEDx German (mTEDx), fully embedded with 16kHz WAV audio bytes, normalized orthographic text, and pre-computed International Phonetic Alphabet (IPA) transcriptions.
📊 Dataset Summary
Total Samples: 1,354 high-quality audio recordings.
Total Size: ~419 MB (Compressed Parquet format).
Audio… See the full description on the dataset page: https://huggingface.co/datasets/q1805/german-golden-audio_speech-IPA.slr72-es-ipa-v1
OpenSLR72 Colombian Spanish — canonical IPA v1
This repository is the approved deterministic v1 publication cut of the OpenSLR72 Colombian Spanish recipe data. It contains the accepted final Lhotse cuts only:
split
rows
train
3,922
test
981
The data files have exactly three columns:
id — the unique final Lhotse cut identifier (for example, com_04310_01996434277-0).
audio — a Hugging Face Audio feature containing the original encoded source bytes and source… See the full description on the dataset page: https://huggingface.co/datasets/bookbot/slr72-es-ipa-v1.common-voice-23-0-es-ipa-v1
Common Voice 23.0 Spanish — canonical IPA v1
This repository is the approved deterministic v1 publication cut of Spanish Common Voice 23.0 recipe cuts. It contains the accepted final Lhotse cuts only:
split
rows
train
352,941
test
15,857
The data files have exactly three columns:
id — the unique final Lhotse cut identifier (for example, common_voice_es_19696062-9131).
audio — a Hugging Face Audio feature containing the original encoded source bytes and source… See the full description on the dataset page: https://huggingface.co/datasets/bookbot/common-voice-23-0-es-ipa-v1.wiktionary-ipa-audio
Wiktionary IPA Audio Mirror
This is a materialized audio mirror of mostol/wiktionary-ipa, published so
users do not need to individually download Wikimedia Commons URLs.
This initial snapshot is partial and contains the successfully downloaded
deduplicated rows available at publication time. metadata.csv maps each
WAV file to its IPA transcription and original source URL. Audio is 16 kHz,
mono, PCM WAV.
The original source URLs are retained for provenance and attribution. Please… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/wiktionary-ipa-audio.Knesset-VOX-IPA
Knesset VOX IPA
Hebrew speech dataset derived from Knesset (Israeli Parliament) plenary sessions, enriched with IPA (International Phonetic Alphabet) phoneme transcriptions. Inspired by the methodology of arxiv:2603.01270.
Dataset Description
Long-form Knesset recordings were split into chunks of up to 15 seconds. Each chunk was transcribed to Hebrew text and then processed for IPA phoneme extraction from audio.
Each sample pairs a WAV audio chunk with:
The original… See the full description on the dataset page: https://huggingface.co/datasets/notmax123/Knesset-VOX-IPA.ghana-speech-ipa-asr-ready
ghana-speech-ipa-asr-ready
The training-ready corpus behind
ghananlpcommunity/ghana-speech-phoneme-asr:
16 kHz audio paired with IPA phoneme targets, already in the partitioned parquet layout
that omnilingual-asr's
MixtureParquetStorage expects.
Use this if you want to resume or repeat the training without redoing ingestion, which
takes a couple of hours of streaming and re-encoding.
Languages
42
Train clips
1,364,950
Dev clips
11,871
Audio
2,329 h, 16 kHz… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/ghana-speech-ipa-asr-ready.Ghana_English-Twi_Code-switching_Speech-ipa
KasaSpeech English–Twi Code-Switching Speech — IPA
A phonemised version of
ghananlpcommunity/Ghana_English-Twi_Code-switching_Speech
(KasaSpeech) with one added column: ipa.
Every other column — audio included — is carried over byte-for-byte, and row
order is unchanged, so this dataset aligns one-to-one with the original.
The ipa column
Each transcript is converted to a phoneme sequence with
ghanag2p-uni, the Twi-only
grapheme-to-phoneme library built on
ghana-g2p… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/Ghana_English-Twi_Code-switching_Speech-ipa.ghana-english-speech-ipa
Ghanaian English Speech — Audio with IPA Transcripts
Speech with both transcript forms: the original orthography and the IPA phoneme
sequence read off the audio by ASR. Each language is a subset, with real
train/validation splits.
from datasets import load_dataset
ds = load_dataset("ghanaopendata/ghana-english-speech-ipa", "English_eng", split="train")
ds[0]["audio"] # decoded waveform, 16 kHz
ds[0]["text"] # original orthography
ds[0]["ipa"] # IPA phonemes
52,855… See the full description on the dataset page: https://huggingface.co/datasets/ghananlpcommunity/ghana-english-speech-ipa.ipa-phonebpe-22lang
22-Language IPA Lexicons and PhoneBPE Vocabulary
This repository contains IPA pronunciation lexicons for 22 languages and a shared multilingual PhoneBPE vocabulary trained on the phoneme transcriptions of all 22 language training sets.
Languages
Code
Language
中文名称
ba
Bashkir
巴什基尔语
be
Belarusian
白俄罗斯语
cs
Czech
捷克语
de
German
德语
el
Greek
希腊语
en
English
英语
es
Spanish
西班牙语
fi
Finnish
芬兰语
fr
French
法语
it
Italian
意大利语
ku
Kurdish
库尔德语
ky… See the full description on the dataset page: https://huggingface.co/datasets/maxwellziweiwei/ipa-phonebpe-22lang.volga-ipatava-znak-vialikaga-magistra-valiantsin-aksiantsiuk-output_original
Знак Вялікага Магістра — арыгінальнае аўдыё
Аўтар / Author: Вольга ІпатаваМова / Language: Беларуская (Belarusian)
Арыгінальнае аўдыё без апрацоўкі, захаванае ў зыходнай якасці.
Частка калекцыі Ministerskija —
корпус беларускіх аўдыёкніг.
Апрацаваная версія (сегменты ~15 с, выраўнаваная транскрыпцыя):
volga-ipatava-znak-vialikaga-magistra-valiantsin-aksiantsiuk-output
Доўгасць аўдыё
4h13m
Радкоў у датасеце
1 302
Структура
Кожны радок змяшчае:… See the full description on the dataset page: https://huggingface.co/datasets/fosters/volga-ipatava-znak-vialikaga-magistra-valiantsin-aksiantsiuk-output_original.volga-ipatava-znak-vialikaga-magistra-valiantsin-aksiantsiuk_all
AudioSet Pipeline Output
Мова / Language: Беларуская (Belarusian)
Аўдыё нарэзана з арыгінальнага запісу ў зыходнай частаце дыскрэтызацыі (native), мона, фрагменты да 30 секунд.
Частка калекцыі Belarusian Audiobooks (native).
Радкоў у датасеце
2,373
Працягласць
10 гадз 9 хв
Частата дыскрэтызацыі
22050 Hz
Каналы
мона
Даўжыня фрагмента
да 30 с
Структура
Кожны радок змяшчае:
audio — аўдыёфрагмент (native SR, мона, ≤30 с)
text — транскрыпцыя… See the full description on the dataset page: https://huggingface.co/datasets/fosters/volga-ipatava-znak-vialikaga-magistra-valiantsin-aksiantsiuk_all.