Glossary
Bring your own key (BYOK)
Bring your own key (BYOK) is a setup in which you supply your own API key for an AI or speech provider. The app then sends requests through your account with that provider, so your agreement with the provider governs the data.
Updated October 16, 20265 cited facts, each traced to its source
Also called BYOK, Own API key.
Bring your own key (BYOK) is an arrangement in which an app uses an API key that you created with an AI provider, instead of a key the app's maker holds. An API key is a secret string that identifies an account to a provider's service. With BYOK, the app sends your audio or text to the provider through your account.
The key moves the data under your provider agreement
Superwhisper, a dictation app for Mac and Windows 1, says that with your own keys the app acts as a client of your provider account. Data handling then follows your agreement with that provider, not the app maker's terms 2. In practice, the provider's policy is the one that applies to your audio, so read it before you connect a key.
Two routes: a key or a local model
BYOK is one of two privacy choices in many dictation apps. The other is a local model, which does not send speech to a cloud service. On Superwhisper's Free plan, dictation uses local Whisper models without AI processing, and the company says nothing dictated leaves the device 3. Cloud models and AI processing come with the Pro plan 4. So a local setup needs no key, while a key connects you to a cloud provider.
An example: a clinic with an existing agreement
Suppose a clinic already has a business associate agreement (BAA) with an AI provider. A BAA is a contract that US health privacy rules require between a covered organization and a vendor that handles protected health information. Superwhisper's documentation calls BYOK the recommended path for organizations with existing HIPAA BAAs or financial data compliance agreements 5. The clinic can then keep dictated notes under the agreement it already has. That is the reason BYOK matters to regulated buyers: the data follows a contract the organization already signed.
The vendor's recommendation does not make a setup compliant on its own. The clinic still has to confirm the provider's terms and its own policies.
Why it matters for dictation
Dictation sends speech, which can be sensitive, to a server whenever it uses a cloud model. BYOK changes whose account and whose terms apply. It does not make the audio private by itself, so check where the audio goes, whether the provider keeps it and whether your agreement covers the use. Our guides on whether voice to text is safe and on HIPAA compliant dictation cover those checks in more detail. For the difference between processing on the device and in the cloud, see on-device speech recognition and cloud dictation.
What to check before you connect a key
- Your provider is supported. Apps support different providers, so confirm yours is on the list.
- The provider's data policy. Once you connect a key, that policy applies to your audio, not the app's.
- The plan. Some apps reserve custom keys for paid tiers, so check which plan includes them before you pay.
- Your own agreements. If you work under a BAA or a similar contract, check that the provider agreement covers your use.
For a wider comparison of AI-based dictation apps and how they handle keys and cloud processing, read our guide to AI dictation apps. For the cost side of cloud AI dictation, read how much AI dictation costs. Related terms include AI dictation and perpetual license, which covers how you pay for an app, a question that is separate from who holds the AI account.
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
- 5.
Related terms
- Dictation Software: How to Choose the Right AppThe topic this term belongs to
- AI dictationAI dictation is dictation in which a language model edits the recognized text after speech recognition, removing filler words, fixing punctuation and sometimes rewording a sentence. It can change your meaning, so check the result before you send it.
- Cloud dictationCloud dictation sends your recorded speech to a remote server, which recognizes it and returns the text. It usually needs an internet connection, and your audio leaves your device, so check the provider's policy before you dictate sensitive material.
- Custom vocabularyA custom vocabulary is a list of words, names and phrases that you add to dictation software so it recognizes and spells them as you intend. It matters most for names, jargon and product terms the recognizer has not learned.
- Dictation softwareDictation software is a program that recognizes your speech and types the words where your cursor is, as you say them. It comes built into many phones and computers and is also sold as separate apps.
- Dragon Medical OneDragon Medical One, abbreviated DMO, is Nuance's speech recognition product for clinicians who dictate notes into health records. Microsoft now combines it with DAX Copilot in a product called Dragon Copilot.
- Dragon NaturallySpeakingDragon NaturallySpeaking, usually called Dragon, is a speech recognition program for dictating text on a computer. Dragon Systems released it in 1997, and Nuance has sold it in several editions since.
- Dragon Professional AnywhereDragon Professional Anywhere is a subscription edition of Dragon that Nuance sells for professional and enterprise users. It is often shortened to Dragon Anywhere, a name that also belongs to Nuance's mobile app, so check which product a listing means.
- Free tierA free tier is the no-cost version of a dictation app, with limits such as a weekly word cap, fewer modes or no AI processing. Read those limits before you rely on it, because a free plan can run out in the middle of your work.
- Nuance CommunicationsNuance Communications was a speech technology company best known for the Dragon family of speech recognition products. Microsoft announced that it had completed its acquisition of Nuance on March 4, 2022.
- Perpetual licenseA perpetual license is a software license you pay for once. It lets you keep using that version without renewing. Updates, support and new major versions are often separate, time-limited or paid extra.
- PowerScribePowerScribe is a family of radiology reporting products in which a radiologist dictates an imaging report by voice. It began as a Nuance dictation application, and Microsoft completed its purchase of Nuance in 2022, so Microsoft now sells the line.
- Voice profileA voice profile is the set of files that dictation software keeps for one user, with settings such as the microphone they use and their custom words. It is not the same as speaker recognition, which identifies who is speaking.