Open ASR Leaderboard
PulseAugur coverage of Open ASR Leaderboard — every cluster mentioning Open ASR Leaderboard across labs, papers, and developer communities, ranked by signal.
- 2026-05-20 product_launch Hugging Face added the Benchmaxxer Repellant benchmark to its Open ASR Leaderboard. source
5 day(s) with sentiment data
-
Open ASR Leaderboard adds Egyptian and Nigerian English accents
Hugging Face has expanded its Open ASR Leaderboard by adding new private datasets featuring Egyptian and Nigerian English accents. This collaboration with Liva AI aims to improve representation and trust in automatic sp…
-
Nvidia releases open-weight Nemotron 3.5 Transcribe model
Nvidia has launched Nemotron 3.5 Transcribe, an open-weight model capable of transcribing audio in 40 languages and identifying up to eight speakers. This release, along with an updated open-source speech dataset called…
-
NVIDIA Nemotron 3.5 Transcribe and Granary v2 offer advanced multilingual speech processing
NVIDIA has released Nemotron 3.5 Transcribe and Granary v2, an open-weight model and data pipeline for advanced multilingual speech processing. Nemotron 3.5 Transcribe offers transcription, speaker attribution, translat…
-
80,000 Hours seeks legal counsel to navigate AI risks and policy
80,000 Hours, an organization focused on mitigating AI risks, is seeking a Head of Legal/General Counsel to manage its legal strategy and operations. The role, based in London or remote, involves advising on novel legal…
-
IBM releases Granite Speech 5.0 2B, ranking #2 on Open ASR Leaderboard
IBM has launched Granite Speech 5.0 2B, a new 2-billion parameter open-source speech-language model designed for multilingual transcription and translation. This model achieves the second-highest ranking on the Open ASR…
-
IBM updates Granite Speech models for improved accuracy and speed
IBM has released updates to its Granite Speech collection, introducing a new 2-billion parameter model, granite-speech-5.0-2b, and improved versions of two existing models. The new 2B model offers enhanced accuracy, rob…
-
New CAFE method enhances multilingual ASR accuracy
Researchers have developed CAFE (Confined-Adapter Feature distillation), a novel method to improve multilingual automatic speech recognition (ASR) systems. CAFE addresses a trade-off where feature distillation technique…
-
Hugging Face leaderboards: How benchmarks are populated and viewed
Hugging Face maintains leaderboards for 48 official benchmarks, which are populated by model authors submitting results via .eval_results YAML files in their model repositories. These leaderboards are automatically asse…
-
Hugging Face consolidates 48 AI benchmarks into interactive map
Hugging Face has launched a new interactive map consolidating its 48 official benchmarks, providing a comprehensive overview of AI model performance across various domains. The map reveals that while agent-related bench…
-
Phonon-2: Small, Accurate Open-Source Speech Recognition Model Released
FermionResearch has released Phonon-2, an open-source speech recognition model that achieves high accuracy for English audio with a small file size. The model averages 5.21% word error rate on the Open ASR Leaderboard, …
-
Hugging Face adds Global South languages to Open ASR Leaderboard
Hugging Face has expanded its Open ASR Leaderboard to include languages from the Global South. This initiative aims to broaden the representation of languages in automatic speech recognition (ASR) technology. The inclus…
-
Production Greek-English Speech Recognizer Developed Through Iterative Engineering
Researchers detailed the extensive engineering process to develop Sophea, a bilingual Greek-English automatic speech recognition system. The team faced challenges in meeting nine production quality gates, particularly b…
-
Hugging Face launches Arabic LLM and ASR leaderboards
Hugging Face has released two new leaderboards: the QIMMA Quality First Arabic LLM Leaderboard, developed in collaboration with the Technology Innovation Institute, and an update to the Open ASR Leaderboard that include…
-
Hugging Face updates open models report, adds Global South language to ASR leaderboard
Hugging Face has updated its State of Open Models report for Summer 2026, detailing the current landscape of open-source AI models. Additionally, the Open ASR Leaderboard has expanded to include its first language from …
-
Hugging Face adds Global South languages to Open ASR Leaderboard
Hugging Face has expanded its Open ASR Leaderboard by introducing two new evaluation sets, Monsoon en-IN and Monsoon hi-IN, to better represent languages from the Global South. These additions aim to address the known i…
-
Hugging Face launches Arabic LLM and ASR leaderboards
Hugging Face has released two new leaderboards: the Qimma Quality First Arabic LLM Leaderboard, developed in collaboration with the Technology Innovation Institute, to evaluate Arabic large language models, and the Open…
-
Hugging Face updates Open ASR Leaderboard and NVIDIA's Nemotron 3.5 safety features
Hugging Face has announced updates related to AI safety and performance. The Open ASR Leaderboard has been updated to include Benchmaxxer Repellant, a tool for evaluating automatic speech recognition systems. Additional…
-
Hugging Face hosts Arabic LLM leaderboard, ASR updates, and NVIDIA safety model
Hugging Face is involved in multiple AI-related announcements. The QIMMA leaderboard focuses on Arabic LLMs, emphasizing quality. Separately, the Open ASR Leaderboard has been updated with Benchmaxxer Repellant. Additio…
-
Hugging Face highlights AI openness, Arabic LLMs, and ASR tools · 3 sources tracked
Hugging Face is highlighting several initiatives related to AI and its applications. One blog post discusses the importance of openness in AI and cybersecurity, mentioning OpenAI. Another post introduces the QIMMA leade…
-
Hugging Face announces updates including Transformers.js, new leaderboards, and AI tools
Hugging Face has released several updates and new projects across its platform. These include Transformers.js v4 available on NPM, custom CUDA kernels for Codex and Claude, and the Open ASR Leaderboard with new multilin…