Qwen-Audio-3.0-ASR

Hear more.
Miss less.

A large-scale, instruction-controlled ASR model for multilingual speech, 16 Chinese dialectal varieties, long-context transcription, and production-grade entity recognition.

Explore audio Inspect results ↓
30
languages
16
Chinese dialects
3B / 35B
active / total params
Overview of Qwen-Audio-3.0-ASR and its production-oriented capabilities
Figure 1 One instruction-controlled model for multilingual speech, streaming, context, hotwords, entities, and polishing.

Reported evaluation

One model,
balanced performance.

9.40%

Macro CER on the internal 16-dialect evaluation suite.

4.57

Macro error rate on Common Voice 15 across the evaluated languages.

11/15

Industry domains where the model reports the outright highest entity recall, with one additional tie.

Interactive audio atlas

Listen across dialects and languages.

Choose a location on either map to hear the original sample and read its source-language transcript with an English translation.

30supported languages
Select a location

Japanese · Japan

日本語

Choose a marker on the map or a name below.

Flexible language prompting

Combine target languages at runtime.

language_hints = ["zh", "ja", "en"]

Browse all 30 supported languages
CER across 16 Chinese dialects
Figure 6 · Recognition accuracy across the same 16 dialect varieties.
Dialect consistency across 16 Chinese dialects
Figure 7 · ASR and AST task-consistency rates.

Fast & streaming

Listen while the transcript appears.

Play the 18-second Beijing-to-Hangzhou request and watch the transcript appear continuously in a clean side-by-side streaming interface.

Interactive streaming playbackDoubao Voice InputvsQwen-Audio-3.0-ASR
Beijing → Hangzhou travel requestVoice-timbre adjusted · original comparison replay
0:00/0:18
DOUBAO VOICE INPUTQWEN-AUDIO-3.0-ASR

Evaluation evidence

Results, where they support the story.

Each result view answers a specific performance question; the complete reported tables remain available below. Lower CER/WER is better, while higher recall and consistency are better.

CER comparison across 16 Chinese dialect test sets

Evaluation note. Dialect and industry-domain evaluations use internal test sets. Public benchmark results in Panel A are taken from official sources, while API systems in Panel B use a unified evaluation pipeline.

Table 1 Public Chinese and English ASR benchmarks
Table 2 Multilingual benchmark results
Table 3 Hotword recall with and without conditioning
Table 4 Long-audio contextual corrections

Samples from the report and launch material

See what the controls change.

Explore context corrections, entity recall, hotword gains, and native polishing without leaving the page.

Model experience

Choose the deployment profile.

Open the official Model Studio documentation for short-form, file-transcription, or real-time streaming integration.

01 / Short-form

Qwen-Audio-3.0-ASR-Flash

Speech recognition for audio up to five minutes.

Open documentation ↗ 02 / File

Qwen-Audio-3.0-ASR-Flash-Filetrans

Offline transcription for recorded audio files.

Open documentation ↗ 03 / Streaming

Qwen-Audio-3.0-ASR-Flash-Streaming

Real-time speech recognition over WebSocket.

Open documentation ↗