# Hugging Face — Open ASR Leaderboard: Trends and Insights with New Multilingual & Long-Form Tracks

- Company: Hugging Face (huggingface.co)
- Announced: 2025-11-21T00:00:00+00:00
- Category: research-paper
- Subject: Platform
- Models affected: Canary-Qwen-2.5B, Granite-Speech-3.3-8B, Phi-4-Multimodal-Instruct, Whisper Large v3, Distil-Whisper, CrisperWhisper, Massively Multilingual Speech (MMS), Omnilingual ASR, Parakeet CTC 1.1B, Parakeet, Voxtral, Whisper
- Source: https://huggingface.co/blog/open-asr-leaderboard
- Record: https://forck.live/items/1583-open-asr-leaderboard-trends-and-insights-with-new-multilingual-long-form-tracks

The Open ASR Leaderboard has added multilingual and long-form transcription tracks. A preprint analyzing 60+ models from 18 organizations across 11 datasets reveals that Conformer encoder + LLM decoder models achieve the best accuracy, CTC/TDT decoders are the fastest, multilingual models sacrifice single-language performance, and closed-source systems still lead in long-form transcription.

## Evidence

Verbatim from https://huggingface.co/blog/open-asr-leaderboard:

> Over the past two years, the Open ASR Leaderboard has become a standard for comparing open and closed-source models on both accuracy and efficiency. Recently, multilingual and long-form transcription tracks have been added to the leaderboard 🎉

---

Record: https://forck.live/items/1583-open-asr-leaderboard-trends-and-insights-with-new-multilingual-long-form-tracks
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md