aTrain Transcription Activity
(THIS PAGE IS UNDER CONSTRUCTION!)
aTrain is a transcription software that uses open-source AI models to transcribe audio files, runs locally on your computer, and therefore does not rely on any cloud-based services. In addition, aTrain can detect speakers. When using aTrain, you can choose which model you want to use to transcribe your file: simpler models will transcribe faster but might make more mistakes, while larger models are more accurate but take more time. If you want to know about the differences in error rates and run time for each model, or about aTrain in general, check out their research paper. In fact, if you are using aTrain for research, you should cite the paper above.
To transcribe a file using aTrain, follow the steps below.
-
If you are using Windows, download aTrain from the Microsoft Store. You won’t need admin permission to download aTrain, but if you have Microsoft Store disabled for some reason, you can also download it directly from here. If you are on a Mac, you can also download aTrain from this link, but it will only work on Macs that use Apple Silicon processors.
-
Open the aTrain app.
Congratulations! You’ve exported your file, which is now ready to edit/clean in Microsoft Word or your text editor of choice.