Frequently Asked Questions

Waveform Studio is an open source audio transcription workspace. It lets you transcribe audio files using AI (powered by Whisper), then view and edit the resulting transcript alongside an interactive waveform with time-aligned highlighting. You can manage speakers, edit segments, and export to a variety of formats.

The source code is freely available on GitLab. You can self-host it, or use the hosted cloud service at this site.

The source code is open source and free to use under its licence. You can self-host the full application on your own infrastructure at no cost.

The hosted cloud service (this site) is a paid option that provides AI transcription and cloud storage without any setup. It is currently invite-only — click Create Account on the home page to request access.

The standalone desktop app is also free to download and run locally, with no account required.

Good news! You don't! There are many free options including running the sourcecode yourself. See Is it free to use for more details.

In order to provide you with an alternative to self-hosting, we have created the Waveform Studio webapp, which utilizes the exact same underlying technology but is web hosted. This means that all of your data is stored safely in the cloud, and gives you the option to share, present and embed, as well as use powerful GPU-optimized transcription.

Unfortunately, web hosting costs us money. Things like storage, transcription processing, and API access require server maintenance, and often pay-as-you-go SaaS costs, which can all add up very quickly. The general rule of thumb is that on the largest Whisper model, 1 hr of audio will take 1 hour to process, and the cloud processing takes around $0.60 USD per hour. Also, storage costs around $0.03 per gigabyte every month. Our pricing model takes those rates into account, as well as adding a standard 30% markup to pay the developers (we do need to eat after all). We have also included a free tier which allows a small amount of storage and transcription processing per month, allowing you to try out the product or simply use it once in a blue moon. Or a normal moon, as the case may be...

Waveform Studio will always be free and open source, but to alleviate the hassle of running it yourself, as well as bringing you a growing suite of tools that are not possible without a central cloud service, we give you the option to join our hosted service at a fair price needed to pay for services and support the developers.

This software uses the AGPL 3.0 license with the Commons Clause. These are both standard licenses that are designed for the type of hybrid open-source project that Waveform Studio is. There's a lot of legalese to sort through, but at a glance, it boils down to the following.

You have free license to utilize this software and the content uploaded to or created by the software for any purpose, including monetization. You have free license to "fork" or modify the software's source code, with the stipulation that in doing so, that code must be open source and publicly available. What you may not do, is sell, commercialize, or otherwise monetize the source code itself or any derivatives of the source code.

Just to be safe, we want to clarify that the license description in this FAQ is a non-legally binding summary of the full license, which can be seen in its entirety here.

Yes. The full source code is available on GitLab. You can deploy the app on your own server to get AI transcription, multi-user support, and cloud storage — with complete control over your data.

See the documentation for setup instructions.

Click Create Account on the home page and fill in your details. The hosted service is currently invite-only — you'll receive an email at the address you provided when your account is activated.

If you'd prefer not to wait, the app is open source and free to self-host.

The standalone app is a downloadable Windows executable built from the open source code. It runs entirely on your own machine — no internet connection or account is required. You can open and edit existing transcripts locally.

Note: AI transcription is not yet available in the standalone version. New transcription requires either the hosted cloud service or a self-hosted server with Whisper configured.

Currently Windows 10 and Windows 11 (64-bit) only. macOS and Linux are not yet supported.

Waveform Studio accepts most common audio formats including MP3, WAV, FLAC, OGG, M4A, AAC, WebM, and OPUS. The transcription backend uses Whisper, which handles a broad range of audio quality and languages.

Transcription is powered by OpenAI Whisper, which supports a wide range of languages. You can specify the language before transcribing, or let Whisper detect it automatically.

You can export transcripts as CSV, DOCX, PDF, or plain text. Each format can be configured to include or exclude timestamps, speaker labels, and other metadata.

The full user guide is available at /docs. It walks through everything from creating your first project and loading audio, to editing transcripts, managing speakers, and exporting results.

Full documentation is available at waveformstudio.app/docs. It covers the workspace interface, all panels, editing, speaker management, keyboard shortcuts, and more.