{"slug": "egyptian-and-nigerian-english-join-the-open-asr-leaderboard", "title": "Egyptian and Nigerian English join the Open ASR Leaderboard", "summary": "Hugging Face's Open ASR Leaderboard added two private English speech-recognition test splits from Liva AI covering Egyptian and Nigerian accents, the organization announced. The Liva AI Conversational EG set contains 1.34 hours of spontaneous Egyptian-accented English and the Liva AI Conversational NG set contains 1.55 hours of Nigerian-accented English, both transcribed with punctuation, casing and disfluencies. The leaderboard will now break down results using Kachru's Three Circles of English and plans to split the Outer and Expanding Circles as more accents join, part of an effort to improve accent representation and limit benchmark overfitting.", "body_md": "[🎙    10](https://huggingface.co/spaces/HumeAI/asr-leaderboard)  \n\n#### RW-Voice EQ ASR\n\nExplore and compare English speech‑recognition model performance\n\nA few months ago, we added private data to the [Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard): 11 splits from Appen and DataOceanAI covering scripted and conversational speech from five accents ([blog](https://huggingface.co/blog/open-asr-leaderboard-private-data)). The community response was great, with many **highlighting the importance of held-out sets for trustworthy evaluations**. Since then, we added one more private split for Indian-accented English with Voice Arena ([blog](https://huggingface.co/blog/open-asr-leaderboard-global-south)).\n\nWe’re excited to announce a new collaboration with [Liva AI](https://huggingface.co/liva-ai) to add even more private data with two new accents: **Egyptian** 🇪🇬 and **Nigerian** 🇳🇬.\n\n| Dataset | Accent | Duration [h] | Style | Transcription | \n|---|---|---|---|---|\n| Liva AI Conversational EG | Egyptian | 1.34 | Conversational, spontaneous | Punctuated, cased, with disfluencies. | \n| Liva AI Conversational NG | Nigerian | 1.55 | Conversational, spontaneous | Punctuated, cased, with disfluencies. | \n\nThis is part of our 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). We’re excited to see the community continuing to push ASR performance for the world 🤗\n\nWith these new accents, we can now present a richer breakdown of accents, based on [Kachru’s Three Circles](https://en.wikipedia.org/wiki/World_Englishes#Kachru%27s_Three_Circles_of_English) (kudos to Hume AI for this idea in their [leaderboard](https://huggingface.co/spaces/HumeAI/asr-leaderboard)). The Open ASR Leaderboard will now report:\n\nAs more accents join, we’ll split the Outer and Expanding Circles.\n\nLiva AI collects multimodal interaction data through consumer products used by real people. Because Liva owns both the collection environment and distribution, it can design products around specific model gaps across domains, industries, languages, modalities, and interaction types. This gives it access to natural, high-context behavior that emerges through real product use, rather than artificially elicited or task-based interactions.", "url": "https://wpnews.pro/news/egyptian-and-nigerian-english-join-the-open-asr-leaderboard", "canonical_source": "https://huggingface.co/blog/open-asr-leaderboard-liva", "published_at": "2026-11-06 00:00:00+00:00", "updated_at": "2026-11-06 10:23:06.494112+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "natural-language-processing"], "entities": ["Hugging Face", "Open ASR Leaderboard", "Liva AI", "Hume AI", "Appen", "DataOceanAI", "Voice Arena"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/egyptian-and-nigerian-english-join-the-open-asr-leaderboard", "markdown": "https://wpnews.pro/news/egyptian-and-nigerian-english-join-the-open-asr-leaderboard.md", "text": "https://wpnews.pro/news/egyptian-and-nigerian-english-join-the-open-asr-leaderboard.txt", "jsonld": "https://wpnews.pro/news/egyptian-and-nigerian-english-join-the-open-asr-leaderboard.jsonld"}}