# Hy ASR 3.0 preview launch notes: Voice Benchmark Profile

> Tencent’s Hunyuan site describes Hy ASR 3.0 preview as a Hy3-based recogniser with context-aware correction and broad dialect coverage; launch coverage cites word error rates near 3%.

- Scope: Dialect coverage and context-aware recognition claims from Tencent, plus launch-coverage WER figures
- Measurement lanes: Voice experience
- Primary metric: Word error rate (lower is better)
- Available evidence: Tencent’s site is descriptive; the WER values come from launch coverage and are marked secondary
- Owner: Tencent Hunyuan
- Source snapshots refreshed: 2026-09-11

## Interpretation limit

The WER figures are reported by launch coverage (BigGo, August 5, 2026) rather than a Tencent page BenchLM could retrieve; treat them as secondary.

## Primary sources

- [Owner page](https://hunyuan.tencent.com/)

Canonical page: https://benchlm.ai/voice-benchmarks/hy-asr-3-0-preview-launch-evaluation