FREEDOM251

Freedom251 › Audio Transcription Software

aTrain

Free plan LinuxMacWebWindows

Visit aTrain Alternatives

What aTrain is

aTrain is a free, open-source application that automatically transcribes speech recordings for qualitative research. Processing happens on the user’s device, so recordings and transcripts are not sent to the internet. The software provides faster-whisper, an implementation of OpenAI’s Whisper model, and speaker identification through pyannote.audio. Listed language support is 99 languages, while the BANDAS product page describes processing in 57 languages. Transcription files can be imported into ATLAS.ti and MAXQDA, with timestamps that connect text to the matching audio. aTrain runs on Linux, macOS, Windows, and the web. Its browser/server mode is experimental and may have incomplete functionality. Processing can use a CPU or NVIDIA GPU; GPU use requires the CUDA toolkit. The highest-quality model is estimated to take around three times the audio length on typical current mobile CPUs. The project uses the AGPL-3.0 license.

Who it suits

aTrain suits qualitative researchers who need local speech transcription, speaker identification, and files that connect with ATLAS.ti or MAXQDA.

What it does well

  • Processes recordings offline on the device
  • Speaker identification included
  • Imports into ATLAS.ti and MAXQDA
  • Runs on Linux, macOS, Windows, and web
  • Free and open source

What to know first

  • Browser/server mode is experimental
  • Some browser functionality may be incomplete
  • Highest-quality model can take about three times audio length
  • GPU use requires the CUDA toolkit

Limits

The highest-quality model may take around three times the audio duration on typical current mobile CPUs. Browser/server mode is experimental, and some functionality may be incomplete.

Getting started

  1. Open the aTrain website
  2. Download the build for Windows, Linux or Apple Silicon macOS
  3. Install and launch aTrain on your computer
  4. Choose CPU processing or configure an NVIDIA GPU with the CUDA toolkit
  5. Load a speech recording and select transcription options
  6. Export the timestamped transcription for ATLAS.ti or MAXQDA

Verdict

aTrain provides local, open-source transcription with speaker labeling and research-software imports. It supports several desktop platforms, but browser mode is experimental and higher-quality processing may be slow on mobile CPUs.

Questions

Is aTrain free?

Yes. It is free and open source, with an AGPL-3.0 license.

Which platforms does aTrain support?

It runs on Linux, macOS, Windows and the web. Windows and Linux downloads are listed, along with Apple Silicon macOS builds under versions 1.2.4 and 1.2.2.

Does aTrain work offline?

Yes. Recordings are processed on the user’s device, and recordings or transcripts are not sent to the internet.

Can it identify speakers?

Yes. Speaker detection uses pyannote.audio to associate each text segment with a speaker.

How many languages are supported?

The listed language support is 99 languages, while the BANDAS product page describes processing in 57 languages.

Can files move into qualitative research software?

Yes. Transcription files can be imported into ATLAS.ti and MAXQDA with timestamps for matching audio playback.

Facts from aTrain's own pages

GPU support
aTrain can run on a CPU or an NVIDIA GPU; GPU use requires the CUDA toolkit. · business-analytics.uni-graz.at, 11 Oct 2026
Intended users
The maker describes aTrain as a transcription tool for qualitative research and says researchers developed it. · business-analytics.uni-graz.at, 11 Oct 2026
Languages
The product page says aTrain can process speech recordings in 57 languages. · business-analytics.uni-graz.at, 11 Oct 2026
License
The project’s GitHub repository identifies its license as AGPL-3.0. · github.com, 11 Oct 2026
macOS availability
The download page lists a macOS build for Apple Silicon under version 1.2.4 and 1.2.2. · business-analytics.uni-graz.at, 11 Oct 2026
Notable limitation
The product page estimates that its highest-quality model takes around three times the audio length on typical current mobile CPUs. · business-analytics.uni-graz.at, 4 Oct 2026
Offline processing
It processes recordings on the user’s device and does not send recordings or transcripts to the internet. · business-analytics.uni-graz.at, 11 Oct 2026
Platforms
The project says packaged apps are available through Flathub for Linux and the Microsoft Store for Windows; it also documents pip installation and says macOS is CPU-only. · github.com, 4 Oct 2026
Purpose
aTrain is a tool for automatically transcribing speech recordings for qualitative research. · business-analytics.uni-graz.at, 11 Oct 2026
Research integrations
Its transcription files can be imported into ATLAS.ti and MAXQDA, with timestamps that let users play the corresponding audio segment. · business-analytics.uni-graz.at, 11 Oct 2026
Speaker detection
Its speaker detection mode uses pyannote.audio to identify which speaker each text segment belongs to. · business-analytics.uni-graz.at, 11 Oct 2026
Transcription engine
It provides access to faster-whisper, an implementation of OpenAI’s Whisper model. · business-analytics.uni-graz.at, 11 Oct 2026
Web mode
The project describes a browser/server mode as experimental and notes that some browser-mode functionality may be incomplete. · github.com, 4 Oct 2026
Windows and Linux downloads
The download page lists current downloads for Windows and Linux. · business-analytics.uni-graz.at, 11 Oct 2026

Where aTrain stands on Freedom251

Look at these too