| --- |
| title: BuzzASR |
| emoji: 🐝 |
| colorFrom: yellow |
| colorTo: red |
| sdk: static |
| pinned: false |
| license: mit |
| --- |
| |
| # 🐝 BuzzASR — A Swarm of 100+ Monolingual Speech Recognition Models |
|
|
| **One specialist model per language.** BuzzASR is a suite of **102 monolingual ASR models**, each a fine-tune of |
| [Whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) specialized to a single language. |
|
|
| The work has been accepted at **EMNLP 2026** · built by [Lemn Lab](https://lemn-lab.github.io/buzz-asr/) in collaboration with [EleutherAI](https://huggingface.co/EleutherAI). |
|
|
| ## The idea |
| Big multilingual models spread themselves thin across hundreds of languages. We asked whether a single specialist, |
| the same size as Whisper, could beat the giant generalists. It can to a considerable extent :) |
|
|
| ## Highlights |
| - ✅ Beats Whisper-large-v3 on **89 of 102** languages (FLEURS) |
| - 🏆 **Open-source state-of-the-art** (beats Whisper, Omnilingual 1B/7B, MMS, Qwen3-ASR, Cohere) on **27 languages** |
| - 📉 Cuts character error rate ~3x on average, and 3.45x across Whisper's 51 worst languages |
| - 🧩 Each model is the same 1.55B parameters as Whisper — about 4.5x smaller than the strongest baseline (Omni-7B) |
|
|
| ## A few standouts (combined FLEURS + Common Voice, normalized CER %) |
| | Language | BuzzASR CER | Whisper zero-shot | Note | |
| |---|---|---|---| |
| | [Cantonese](https://huggingface.co/BuzzASR/cantonese) | **13.0** | 36.4 | SOTA — 2x better than the next-best system | |
| | [Punjabi](https://huggingface.co/BuzzASR/punjabi) | **8.8** | 42.3 | SOTA | |
| | [Mongolian](https://huggingface.co/BuzzASR/mongolian) | **5.2** | 38.2 | SOTA | |
| | [Korean](https://huggingface.co/BuzzASR/korean) | **4.6** | 5.7 | SOTA | |
| | [Amharic](https://huggingface.co/BuzzASR/amharic) | **9.1** | 191 | >20x reduction over Whisper | |
|
|
| ## Find your language |
| Browse all 102 models in the **Models** tab above, or go to `huggingface.co/BuzzASR/<language>`. |
|
|
| ## Usage |
| ```python |
| from transformers import WhisperForConditionalGeneration, WhisperProcessor |
| model = WhisperForConditionalGeneration.from_pretrained("BuzzASR/mongolian") |
| proc = WhisperProcessor.from_pretrained("BuzzASR/mongolian") |
| # the language/task prompt is baked in — just call model.generate(input_features) |
| ``` |
|
|
| ## Links |
| - 📄 Paper — Findings of EMNLP 2026 (https://arxiv.org/abs/2609.09554) |
| - 🌐 Project & docs — https://lemn-lab.github.io/buzz-asr/ |
|
|
