🔍 Read the full analysis: AI Speech Recognition Welcomes Egyptian And Nigerian English To Open ASR on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Hugging Face has added two private Liva AI English speech evaluation splits to the Open ASR Leaderboard, covering Egyptian and Nigerian accents. The conversational recordings total 2.89 hours; the announcement does not detail speaker composition or show how the additions affect leaderboard scores.
Hugging Face has added two private English speech evaluation splits from Liva AI to the Open ASR Leaderboard, covering Egyptian and Nigerian accents, as detailed in the original analysis. The 2.89 hours of conversational recordings broaden the accents represented in the leaderboard’s evaluations, though the announcement does not establish how representative the short samples are of speakers in either country.
The new splits are named Liva AI Conversational EG and Liva AI Conversational NG. The Egyptian set contains 1.34 hours of speech and the Nigerian set contains 1.55 hours. Hugging Face describes both as spontaneous conversational audio, with transcripts that are punctuated and cased and retain disfluencies.
The recordings are private evaluation data, not publicly available benchmark datasets. Hugging Face says private splits are intended to make evaluation more trustworthy and reduce developers’ ability to optimize systems against publicly accessible test material. The announcement does not provide a score comparison showing how the new data changes model rankings.
The leaderboard’s Private tab will also show two accent comparisons, according to Hugging Face: US versus non-US English, and Inner Circle versus Outer and Expanding Circles. The non-US grouping includes the UK, Australia, Canada, India, Nigeria and Egypt. The post credits Hume AI with the idea for the circle-based leaderboard comparison and says the framework draws on linguist Braj Kachru’s Three Circles model.
Egyptian and Nigerian English join the Open ASR Leaderboard
Hugging Face added private Liva AI speech splits to broaden accent evaluation. The short recordings add coverage, while leaving questions about how representative the results are.
What was added
Both datasets contain spontaneous conversation with punctuated, cased transcripts that retain disfluencies.
Liva AI Conversational EG
A private evaluation split for Egyptian-accented English.
Liva AI Conversational NG
A private evaluation split for Nigerian-accented English.
New ways to compare accents
The leaderboard’s Private tab will show results across two groupings.
- Non-US includes the UK, Australia, Canada, India, Nigeria and Egypt.
- Inner Circle compared with Outer and Expanding Circles, based on Kachru’s framework.
- Hugging Face plans to separate the latter circles as more accents are added.
Why private evaluation matters
Keeping test examples private may make it harder to tune systems directly to known benchmark material.
Users can inspect recognition results for two additional English varieties.
Scores describe these specific short test sets, not all speakers in either country.
What the New Accent Splits Add
Speech recognition systems can perform differently across accents, so a single overall score may conceal differences in how well a system handles particular speech varieties. Adding Egyptian and Nigerian English gives leaderboard users two additional accent-specific evaluation sets and a way to inspect results through broader regional groupings.
The private status may make it harder for developers to tune models to the exact examples used for evaluation. But these splits are small: they contain less than three hours combined. Their results should be understood as performance on these particular recordings, not as a complete measure of accuracy for all Egyptian or Nigerian English speakers. The announcement does not include independent evidence about how well the samples reflect the wider populations.
Products Worth Considering
Bose QuietComfort Ultra Earbuds (2nd Gen) – Wireless Noise‑Cancelling Earbuds with Mic, Immersive Audio, USB‑C Charging, Up to 6 Hours Batte
Bose QuietComfort Wireless Noise Cancelling Headphones, Black
Bose QuietComfort Headphones (2nd Gen), Wireless Over-Ear Noise-Cancelling Headphones, Black
As an affiliate, we earn on qualifying purchases.
Earlier Open ASR Data Additions
The Liva AI splits continue Hugging Face’s effort to broaden the private data used by the Open ASR Leaderboard. The company says it previously added 11 private splits from Appen and DataOceanAI, covering scripted and conversational speech across five accents. It later added a private split for Indian-accented English with Voice Arena.
The circle groupings classify the US, UK, Australia and Canada as the Inner Circle, while India, Nigeria and Egypt fall within the combined Outer and Expanding Circles grouping in the current presentation. Hugging Face says it plans to separate the latter circles as more accents are added, but has not set a date or specified how many additions would prompt the change.
“We’re excited to announce a new collaboration with Liva AI to add even more private data with two new accents: Egyptian and Nigerian.”
— Hugging Face
What the Samples Do Not Show
The announcement does not say when the recordings were collected, how many speakers took part, or how those speakers were selected. It also gives no breakdown by region, age or other speaker characteristics, leaving readers without information needed to judge how broadly the recordings represent Egyptian and Nigerian English.
Hugging Face has not published baseline scores or described any effect on existing leaderboard rankings in the source material. It is also unclear whether the audio or fuller dataset documentation will ever be made public; the new splits are described as private. Without those details, the scale and practical impact of the change cannot be measured from the announcement alone.
Planned Changes to Accent Groupings
Hugging Face says it plans to keep adding accents and languages and to split the current Outer and Expanding Circles category into separate groupings once more accents are available. No timeline or threshold for that reorganization was provided. For now, users can view the new comparisons in the leaderboard’s Private tab, according to the announcement.
Further information on speaker composition, collection methods and per-accent results would help users interpret the new evaluations. The source material does not state when additional details or future data additions will be announced.
Key Questions
Which English accents are covered by the new splits?
The two private splits cover Egyptian-accented English and Nigerian-accented English.
How much audio do the new evaluation sets contain?
The Egyptian split contains 1.34 hours and the Nigerian split 1.55 hours, for 2.89 hours combined.
What kind of speech is included?
Hugging Face describes the recordings as spontaneous conversational audio. The transcripts are punctuated and cased and include disfluencies.
Can developers download the new datasets?
The splits are described as private evaluation data, not publicly available benchmark datasets. The announcement does not say whether they may be released publicly later.
Do the new splits establish how well speech recognition works for speakers in Egypt and Nigeria?
No. The recordings provide results for these specific test sets. Their short duration and the lack of published details about speaker selection and composition make it unclear how representative they are of speakers more broadly.
Primary source: Hugging Face · via ThorstenMeyerAI.com
