Eric Bezzam, Ashley Mo, Aoi Otani · Nov 6, 2026
Hugging Face post: Egyptian and Nigerian English join the Open ASR Leaderboard
Markdown · JSON · Original source
The post announces two private Liva AI speech data splits for the Open ASR Leaderboard: one for Egyptian English and one for Nigerian English. Each contains spontaneous, conversational speech, with punctuated and cased transcriptions that retain disfluencies. The post presents the additions as part of an effort to represent more accents and languages and to improve trust through private data intended to minimize benchmaxxing and new data providers. The leaderboard will also report accent groupings: US versus non-US, and Inner Circle versus Outer & Expanding Circles, using the categories listed in the post. The authors say they will split Outer and Expanding Circles as more accents join. Escape Meridian interpretation: the post describes new evaluation data and planned groupings, not evidence that any model performs better on these accents. It gives the splits’ durations and transcription style, but does not report evaluation results using them.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
· Article
Source details
Original source date: Nov 6, 2026 (2026-11-06T00:00:00.766Z).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-09T04:47:35.855Z. First observed: 2026-11-06T10:00:23.903Z. Retrieved: 2026-11-09T04:46:52.103Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The post does not report evaluation results using the Egyptian and Nigerian splits, so their effect on reported model performance is unknown. The source describes the data as private and gives its duration and transcription style, but does not provide further dataset details in this article.
Open question
What evaluation results, if any, will the Egyptian and Nigerian splits produce when used on the leaderboard?
Source-stated
Hugging Face announced private Liva AI splits for Egyptian and Nigerian English on the Open ASR Leaderboard.
add even more private data with two new accents: Egyptian 🇪🇬 and Nigerian 🇳🇬.
Source-stated
The two splits contain spontaneous conversational speech, with punctuated, cased transcriptions that include disfluencies.
Conversational, spontaneous Punctuated, cased, with disfluencies.
Source-stated
The post says the additions are intended to improve accent and language representation and trust, including by minimizing benchmaxxing with private data and adding providers.
continued effort to improve representation in ASR evals (more accents and more languages) and to improve trust (new private data to minimize benchmaxxing, and new data providers).
Source-stated
The leaderboard will report US versus non-US and Inner Circle versus Outer & Expanding Circles groupings; the post says it will split the latter as more accents join.
As more accents join, we’ll split the Outer and Expanding Circles.
Escape Meridian interpretation
The post describes the data additions and planned groupings, but does not report evaluation results using the new splits.
With these new accents, we can now present a richer breakdown of accents
Source and contributor context
Eric Bezzam, Ashley Mo, Aoi Otani · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media was declared by the source, but no asset passed the supported availability checks. Open the original source to view it.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Hugging Face
Hugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Editorial provenance
Version 1. Publication ID: 0c8a7b75-cf07-4a69-811d-f16f4d0e2e0d.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T04:47:35.855Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: e217bfbd7afa1dd5e4656ef73ba48e0f1cf50681f92b4f3c0859db42aa731051. Source text hash: 5b2a8c3cd032d092363b1f51d7b9411130664e5de6f859206492fcae60165376.
Revision and source-check history
Version 1 · 2026-11-09T04:47:35.855Z
Hugging Face post: Egyptian and Nigerian English join the Open ASR Leaderboard
The post announces two private Liva AI speech data splits for the Open ASR Leaderboard: one for Egyptian English and one for Nigerian English. Each contains spontaneous, conversational speech, with punctuated and cased transcriptions that retain disfluencies. The post presents the additions as part of an effort to represent more accents and languages and to improve trust through private data intended to minimize benchmaxxing and new data providers. The leaderboard will also report accent groupings: US versus non-US, and Inner Circle versus Outer & Expanding Circles, using the categories listed in the post. The authors say they will split Outer and Expanding Circles as more accents join. Escape Meridian interpretation: the post describes new evaluation data and planned groupings, not evidence that any model performs better on these accents. It gives the splits’ durations and transcription style, but does not report evaluation results using them.
Original source
· Article
Source details
Original source date: Nov 6, 2026 (2026-11-06T00:00:00.766Z).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-09T04:47:35.855Z. First observed: 2026-11-06T10:00:23.903Z. Retrieved: 2026-11-09T04:46:52.103Z.
Limits
The post does not report evaluation results using the Egyptian and Nigerian splits, so their effect on reported model performance is unknown. The source describes the data as private and gives its duration and transcription style, but does not provide further dataset details in this article.
Open question
What evaluation results, if any, will the Egyptian and Nigerian splits produce when used on the leaderboard?
Source-stated
Hugging Face announced private Liva AI splits for Egyptian and Nigerian English on the Open ASR Leaderboard.
add even more private data with two new accents: Egyptian 🇪🇬 and Nigerian 🇳🇬.
Source-stated
The two splits contain spontaneous conversational speech, with punctuated, cased transcriptions that include disfluencies.
Conversational, spontaneous Punctuated, cased, with disfluencies.
Source-stated
The post says the additions are intended to improve accent and language representation and trust, including by minimizing benchmaxxing with private data and adding providers.
continued effort to improve representation in ASR evals (more accents and more languages) and to improve trust (new private data to minimize benchmaxxing, and new data providers).
Source-stated
The leaderboard will report US versus non-US and Inner Circle versus Outer & Expanding Circles groupings; the post says it will split the latter as more accents join.
As more accents join, we’ll split the Outer and Expanding Circles.
Escape Meridian interpretation
The post describes the data additions and planned groupings, but does not report evaluation results using the new splits.
With these new accents, we can now present a richer breakdown of accents
Source and contributor context
Eric Bezzam, Ashley Mo, Aoi Otani · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media was declared by the source, but no asset passed the supported availability checks. Open the original source to view it.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Hugging Face
Hugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Editorial provenance
Version 1. Publication ID: 0c8a7b75-cf07-4a69-811d-f16f4d0e2e0d.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T04:47:35.855Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: e217bfbd7afa1dd5e4656ef73ba48e0f1cf50681f92b4f3c0859db42aa731051. Source text hash: 5b2a8c3cd032d092363b1f51d7b9411130664e5de6f859206492fcae60165376.
uncertain · 2026-11-09T04:47:35.855Z · anthropic/claude-opus-5.5
Publication: 0c8a7b75-cf07-4a69-811d-f16f4d0e2e0d
attribution — pass: The reading correctly attributes the announcement to the Hugging Face blog post, written with Liva AI collaborators, and keeps the post's goals, such as minimizing benchmaxxing, attributed to the post rather than stated as fact.
We’re excited to announce a new collaboration with Liva AI to add even more private data with two new accents
summary — uncertain: The summary is faithful, but key terms are never explained for a general reader. ASR (automatic speech recognition), benchmaxxing, splits, disfluencies and Kachru's Inner/Outer/Expanding Circles are left undefined. The summary also says it uses 'the categories listed in the post' without listing which countries fall in each group. It mentions durations without giving them (about 1.34h and 1.55h), and the small size matters for interpretation. Smallest repair: briefly gloss ASR and benchmaxxing, and state the circle groupings and hours.
Inner Circle (US, UK, Australia, Canada) vs. Outer & Expanding Circles (India, Nigeria, Egypt)
date — pass: The originalPublished date of 2026-11-06 matches the post's stated publication date of November 6, 2026.
Published November 6, 2026
support — pass: Each claim matches its excerpt. The interpretation that the post contains no evaluation results is labeled and consistent with the text. The question does not presuppose any outcome, and the limitations are accurate.
new private data to minimize benchmaxxing, and new data providers
media — pass: The selected media is the verified direct link to the original Hugging Face post, which is the most relevant access point.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Hugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Explore Hugging FaceHugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Explore Hugging FaceHugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Explore Hugging FaceHugging Face hosts models, datasets and software that let people share, adapt and inspect AI systems.
Explore Hugging Face