Qwen3-ASR-0.6B-hf

Name: Qwen3-ASR-0.6B-hf
Author: Qwen

by Qwen

782M multilingual speech recognition, small enough to run over a whole archive

144Kdl/month

63likes

HuggingFace Run it on your own data, free

Identifier

Model ID

Qwen/Qwen3-ASR-0.6B-hf

Overview

Speech recognition is the step that makes audio and video searchable at all. Nothing downstream works until the words exist as text, and the cost of that step decides whether you transcribe everything or only the files someone asks about.

At 782M parameters this is a small model by current standards, which is the interesting property. Whisper-large-class models are more accurate and considerably more expensive per hour of audio; a model this size changes the arithmetic on transcribing an entire back catalogue rather than a sample of it.

The card declares Chinese, English, Cantonese, Arabic, German, French, Spanish and Portuguese, so it is genuinely multilingual rather than English with extras.

Architecture

Qwen3ASRForConditionalGeneration, model type qwen3_asr, 782,426,112 parameters. Automatic-speech-recognition pipeline. Declared languages: zh, en, yue, ar, de, fr, es, pt.

Key Capabilities

•Transcription across eight declared languages
•Small enough for bulk transcription rather than sampling
•Text output that a keyword index can match directly
•A cheap first pass ahead of a larger model on difficult audio

Use Cases on Mixpeek

•Making a podcast or video archive keyword-searchable end to end
•Producing the BM25 side of hybrid search over spoken content
•Timestamped transcripts so a search result points at a moment, not a file
•Multilingual libraries where one English-first model leaves gaps

Use Qwen3-ASR-0.6B-hf on Mixpeek

Build multimodal processing pipelines with this model and others. Extract features, run inference, and set up retrieval in Mixpeek Studio.

Open Studio

How It Runs on Mixpeek

On Mixpeek, Qwen3-ASR-0.6B-hf runs as a managed extractor inside a processing pipeline. Point a bucket of automatic speech recognition data at it, and Mixpeek handles GPU provisioning, batching, retries, and writing the outputs into a vector store you can query.

Extractor outputs land in the Mixpeek Vector Store (MVS), where you can combine them with retrieval, reranking, and filter stages to build end-to-end search and agent-perception pipelines, no model-serving infrastructure to maintain.