@aigclink: Alibaba Tongyi Lab just dropped Fun-ASR 1.5—one industrial-grade model handles 30 languages, all 7 major Chinese dialect families + 20+ regional accents, even classical-poetry recitation. Dialect CER down 56.2 % vs last gen; 5 dialects top 90 % accuracy…
Summary
Alibaba Tongyi Lab releases Fun-ASR 1.5: a single model covering 30 languages, seven Chinese dialect groups and 20+ local accents; character-error rate in key dialect scenarios falls 56.2 %, with five dialects exceeding 90 % accuracy.
View Cached Full Text
Cached at: 04/21/26, 08:26 AM
Alibaba Tongyi Lab has just released a new ASR model: Fun-ASR 1.5
Core capability: production-grade dialect support
A single model seamlessly covers 30 languages, all seven major Chinese dialect families, and 20+ regional accents; even classical-poem chanting is transcribed accurately.
Typical-dialect CER drops 56.2 % relative to the previous generation; five dialects exceed 90 % accuracy, and fifteen surpass 80 %.
Similar Articles
@MaxForAI: If you are working on voice agents, you should try this project. A team from NTU, NUS, and Shanghai AI Lab released: Mega-ASR. This fully open-source ASR is built on Qwen3-ASR, aiming to break the long-standing bottleneck of ASR performance in noisy, reverberant, or other impaired real-world environments...
NTU, NUS, and Shanghai AI Lab jointly released Mega-ASR, a fully open-source ASR model built on Qwen3-ASR. Using the Voices-in-the-Wild-2M dataset and progressive acoustic-to-semantic optimization, it achieves up to 30% relative Word Error Rate (WER) reduction in real-world noisy environments. With only 1.7B parameters, it enables efficient inference on consumer-grade hardware.
@lxfater: NetEase Youdao open-sourced ZiYue 4 model, within 27B parameters, SOTA in math and science. But what really interests me is its voice feature!! Cloning a voice is nothing new, ElevenLabs could do it long ago. But they all share a common flaw: cross-language accent. Take your Chinese voice and use it to speak Japanese — it has a Chinese accent, you can tell it's a foreigner struggling...
NetEase Youdao open-sourced the ZiYue 4 model with 27B parameters, achieving SOTA in math and science; its voice feature supports 3-second cross-language voice cloning across 14 languages with no accent issue, along with open-sourcing the all-scenario intelligent agent 'Longxia' (Lobster).
@seclink: Xiaomi is on a strong growth trajectory! MiMo v2.5-ASR Released on 2026-06-02
Xiaomi has released updates to its MiMo model series, including mimo-v2.5-asr (supporting multiple dialects and lyric transcription), mimo-v2.5-pro (trillion parameters, 1M context), mimo-v2.5 (full-modal perception), and a TTS series, significantly improving agent performance and recognition capability in complex acoustic scenarios.
@FeitengLi: Actually, these problems can be well solved: 1. Ditch whisper, switch to an ASR model. Qwen3-ASR is great with few hallucinations, and there are other ASR options. Whisper has many hallucinations and requires 30s segments. Qwen3-ASR gets more accurate with longer audio, supporting up to 20…
Recommends using Qwen3-ASR instead of Whisper to reduce hallucinations, using LattifAI tools for precise audio-text alignment and subtitle generation, and introducing their own OmniVAD-Kit project for voice activity detection.
@FinanceYF5: OpenAI's new voice model Bidi 1 first test exposure - Bidirectional voice design: while you speak, it listens; you can interrupt mid-way to switch tasks immediately, no longer grabbing the conversation when you pause. It also supports real-time translation, and context memory is much stronger than the current Advanced Voice. It's now being pushed to a small group, ChatGPT …
OpenAI's new voice model Bidi 1 first test exposure, supports bidirectional voice design, real-time translation, and stronger context memory, currently being pushed to a small group on ChatGPT.