attunz help
Using memory for names and technical terms
Attunz builds memory automatically around associated people, projects and input contexts, rather than requiring a hand-maintained vocabulary list. Enable Memory, choose an available memory model and dictate as usual. Review saved memories in Memory Galaxy; document import is a separate action you choose.
When memory can help
Names and specialist terms often have a particular spelling. Memory can associate vocabulary with the people, projects and contexts it belongs to and help speech recognition and AI enhancement use relevant information. This is not a guarantee that a term will be learned or transcribed correctly on every attempt; check important wording.
A practical workflow
- Enable Memory, choose an available memory model and dictate as usual. Attunz extracts, removes duplicates and organizes useful terms in the background for later speech recognition and AI enhancement.
- Read the resulting text and correct any names or technical terms that are wrong.
- Open Settings → Memory and use Memory Galaxy to view your memories.
Checking the result remains useful even when you use memory. If a term is still wrong, correct the text you are about to use; do not assume it will be right on the next attempt.
Memory models and document import
If memory processing is enabled, open Memory settings and select an available local Memory model for offline use. Follow AI Enhancement requires AI Enhancement to be enabled and its selected model to be local for this setup. To keep AI enhancement off while using memory, select an available local memory model explicitly. Choosing local ASR does not change the memory model.
To add your own documents, open Memory in Settings and choose Import Memories. You choose the documents to import. This is separate from automatic dictation context and does not mean that attunz scans your project files automatically.
All Attunz software features are free forever with local models or BYOK (your own API key). Attunz charges no fees for these options. You pay any third-party model fees directly to the provider. The Attunz hosted cloud service uses manual usage-credit top-ups for optional cloud ASR and AI enhancement, with no automatic renewal at present. It is available. Pricing
Speech recognition and AI enhancement
Speech recognition (ASR) turns audio into text. AI enhancement works on that text when enabled. Memory affects both stages and helps improve the accuracy of recognized and enhanced text. It does not remove the need to check the final wording.
During dictation, attunz can automatically use the app name, AX window text, selected text, existing input-field text and available caret position as needed. Window text comes from macOS Accessibility (AX), not screenshots. Available text and caret information depend on the app and permissions; they help AI enhancement understand the input scenario and refine ASR text and formatting.
When enabled and permitted, local Claude/Codex conversation history can also contribute coding context. This is separate from AX window text and documents you deliberately import; it is not automatic scanning of project source files. History stored locally does not by itself determine whether the selected processing model is local or cloud-based.
Clipboard capture for model context is off by default. Automatic window context does not scan project source files. Importing documents into memory is a separate action you choose; permitted local Claude/Codex history is another distinct input. You can turn off automatic context collection.
An example with a project name and a technical term
Hypothetical project-name example: “Write a release note for the Attunz project.” Attunz is the example name whose spelling you would check in the resulting text. This illustrates associated project vocabulary, not an observed ASR output, a before-and-after result or a recorded demo.
What to keep in mind
Automatic memory does not guarantee that every unfamiliar name or term will be correct. Recording conditions and the models you choose can also affect the result. For local model setup and the distinction between local and cloud processing, read the offline dictation guide.
Common questions
How do I add custom vocabulary in attunz?
Enable Memory and choose an available memory model. Attunz builds associated memory automatically, so you do not need to maintain a word-by-word vocabulary list. You may also deliberately import documents. Review saved memories in Memory Galaxy and check the resulting text; automatic memory does not guarantee the next recognition result.
Where can I see the memories?
Open Settings → Memory and use Memory Galaxy to view your memories.
Does memory guarantee correct names and technical terms?
No. Memory helps with speech recognition and AI enhancement, but you should still check important names, terms and the final wording.
What should I do if dictation gets a name or technical term wrong?
Correct the text before you send or publish it. Check that Memory is enabled and an available memory model is selected, then use Memory Galaxy to review the saved terms. You can import a document containing relevant vocabulary through Import Memories. Memory does not guarantee a correct result every time, so check the next output too.
Can I import my own documents into memory?
To add your own documents, open Memory in Settings and choose Import Memories. You choose the documents to import. This is separate from automatic dictation context and does not mean that attunz scans your project files automatically.
Switching from another dictation app? Compare attunz as a Wispr Flow alternative or a Superwhisper alternative.
macOS 14.0+ · Apple Silicon · compatibility requirements