​Audio Transcription and Recording Tool
Version 1.0.0 - Free and Open-Source Software​
To use the tool, please read the installation and model instructions below carefully.
High-Accuracy Audio Transcription
-
Convert your .mp3 and .wav files into text effortlessly.
-
Powered by OpenAI Whisper, ensuring state-of-the-art transcription quality.
-
Supports 8 major languages, including Portuguese, English, Spanish, French, German, Italian, Russian, and Chinese.
Fast Processing with GPU Acceleration
-
Optional GPU support for lightning-fast transcriptions, saving your precious time.
Flexible Whisper Model Options*
-
Choose the model that fits your needs and resources:
Tiny: Super fast, minimal resource usage.
Base: Balanced speed and accuracy.
Small: Reliable accuracy for general transcriptions.
Medium: High accuracy for clear audio.
Large-v3: Maximum accuracy for complex tasks.
Built-In Audio Recorder
-
Record audio directly within AudioT with Start, Pause, and Stop controls.
-
Save recordings as .wav or .mp3 files, ready for transcription or sharing.
-
Experience visual feedback with the blinking record button for active recording.
Instant Transcription Access
-
Save transcriptions as .txt files with one click.
-
View texts in a resizable window with easy copy-to-clipboard functionality.
-
Quickly locate your files with the Folder button.
Multilingual & User-Friendly Interface
-
Switch between 8 interface languages seamlessly.
-
Supports light and dark modes for your comfort.
-
Stay updated with real-time progress bars.
-
Built with robust error handling for smooth, worry-free usage.
Lightweight & Open-Source
-
Works efficiently on Windows devices.
-
Fully open-source, allowing you to trust, customize, and improve it.
Instructions:
1 Preferably, install and run as administrator.​
​​​
2 In this version, the “small.pt” and “medium.pt” multilingual model files are included. They are suitable for most cases and offer the best cost-benefit in terms of speed and accuracy. If you need a different model, you can download it below and manually place the *.pt file into the installation folder at C:/.../AudioT/_internal/whisper/models/. Alternatively, log in to Hugging Face at https://huggingface.co/login, search for Whisper models, and download the desired models from their details page.
​
3 Saving as .mp3 generates a much smaller file size than .wav.
"AudioT simplifies your workflow, whether you’re a student, researcher, journalist, or professional needing accurate and fast audio-to-text conversion. It is a great tool to use in combination with your preferred AI tool."
Alexandre Oliveira, Developer
Light mode.
Dark mode
(Transcripted text goes here...)
Window with transcripted text and copy-to-clipboard button
If AudioT makes your work easier, please consider making a small donation. Your support keeps it free for everyone and helps it grow. Every contribution makes a difference, and it only takes a few seconds!
MediaFire
Mega
Download Whisper Models
MediaFire
Mega