Best Voice to Text Apps for Mac

The short answer. Separate dictation from transcription first, because they are different jobs. If you are talking and want the words to appear where your cursor is, that is dictation: OpenDictation if you want the engine to be a choice rather than a fixed decision, Presspeech if you want push-to-talk with nothing leaving the machine, Aiko if you want on-device transcription without configuring a model, and Google AI Edge Eloquent if you want a free on-device option and work in English. If the audio is a file, a call or a meeting rather than your own voice, that is transcription: LocalScribe for files and system audio, LokalBot when you want the result to become searchable notes rather than text in the current field, and BeMyEars when you need live captions in more than one language.

This guide covers 9 voice to text apps for macOS. Every fact below is drawn from the same verified records that back each app’s own page. The recommendations are our editorial judgment, not payment.

Quick comparison

What actually decides this

• Dictation or transcription. Dictation puts words at the cursor. Transcription turns a recording into a document. The tools overlap in name and almost never in workflow, and choosing the wrong kind is the most common mistake in this category.
• Whether the audio leaves the machine. A local model keeps a meeting, a medical note or a private call on the Mac. A cloud service does not. Several apps here are explicit about running on-device, and that is the property to check first if the content is sensitive.
• Whether the model is a choice. A fixed engine is simpler and gives you one answer. A swappable one lets you trade speed against accuracy, or pick a model that handles your language. Neither is better; the difference is whether you expect to tune it.
• Where the text ends up. Text at the cursor disappears into whatever you were doing. Text into a notes store is searchable later. If you record meetings, the second is what you actually want, and it is a different application.
• Language coverage. Several engines here are strong in English and narrower elsewhere, and the record usually says so. If you dictate in more than one language, check this before anything else.

Our picks

• OpenDictation — Its record describes a local Whisper default with optional engines alongside it, and both press-to-start and hold-to-talk modes. Choose it when you want a dictation utility where the speech engine is a choice rather than a fixed decision.
• Presspeech — Its record describes a local engine with no cloud and no telemetry, and push-to-talk as the interaction model. Choose it when you want push-to-talk dictation and nothing leaving the machine.
• LocalScribe — Its stated scope is microphones, media files and Mac audio together, with a subtitle editor — which is transcription rather than dictation. Choose it when the audio is a file or a call rather than your own voice.
• LokalBot — Its record describes on-device meeting notes and searchable work memory, which is a notes workflow rather than a dictation one. Choose it when you want the transcript to become searchable notes rather than text in the current field.
• Aiko — Its record describes on-device transcription as the product, with no model configuration to make. Choose it when you want on-device transcription without configuring a model.
• BeMyEars — Its record describes live captioning with on-device transcription and multilingual support. Choose it when you need live captions and more than one language.

The apps

Aiko

Aiko is a transcription app for macOS. High-quality on-device transcription. It is free.

What it does. AI-powered audio transcription. High-quality on-device transcription. Easily convert speech to text from meetings, lectures, and more.

Aiko is listed with current version 1.3.0.

Full profile and verified facts for Aiko →

Azex Speech

Azex Speech is a transcription app for macOS. Voice input tuned for mixed Chinese-English dictation in AI and crypto work. It is open source.

Full profile and verified facts for Azex Speech →

BeMyEars

BeMyEars is a transcription app for macOS. Live captioning tool with on-device transcription and multilingual support. It is free.

Full profile and verified facts for BeMyEars →

buzz

buzz is a transcription app for macOS. Transcribes and translates audio offline on your personal computer. Powered by OpenAI’s Whisper. It is released under MIT.

What it does. Transcribe audio and video files or Youtube links. Live realtime audio transcription from microphone. Speech separation before transcription for better accuracy on noisy audio. Speaker identification in transcribed media.

buzz is listed with current version 1.4.5.

Full profile and verified facts for buzz →

Google AI Edge Eloquent

Google AI Edge Eloquent is a productivity app for macOS. Free on-device AI dictation and voice editing from Google that works in any app, currently English only. It is free.

Full profile and verified facts for Google AI Edge Eloquent →

LocalScribe

LocalScribe is a productivity app for macOS. Open-source local transcription and subtitle editor for microphones, media files, and Mac audio. It is released under MIT.

What it does. Apple Speech live captions for microphone, Mac audio, or mixed input on macOS 26. whisper.cpp Metal inference with bundled Silero VAD and automatic CPU fallback. Downloadable Whisper, SenseVoice, and Parakeet model choices. Adaptive progressive text display for file and microphone transcription.

LocalScribe is listed with current version 1.7.0.

Full profile and verified facts for LocalScribe →

LokalBot

LokalBot is a productivity app for macOS. On-device meeting notes, dictation, inline autocomplete, and searchable work memory for macOS. It is released under GPL-3.0.

What it does. LokalBot is a free, open-source Mac app that finds what you said or saw: bot-free meeting notes, on-device transcription, saved screen text, and dictation.

LokalBot is listed with current version 0.9.6.

Full profile and verified facts for LokalBot →

OpenDictation

OpenDictation is a productivity app for macOS. Open-source dictation utility with local and cloud speech-to-text. It is released under MIT.

What it does. Whisper runs locally and remains the default speech engine. Parakeet TDT v2 handles English, while Parakeet TDT v3 supports 25 European languages. Both are optional and run locally through FluidAudio 0.15.6. The global shortcut supports press-to-start/press-to-stop and hold-to-talk modes. Microphone choices use stable Core Audio device identifiers. If a selected device is missing, recording stops with a clear error instead of changing inputs.

OpenDictation is listed with current version 1.2.1.

Full profile and verified facts for OpenDictation →

Presspeech

Presspeech is a productivity app for macOS. Native push-to-talk local dictation app for Apple Silicon Macs using Parakeet TDT v3 (CoreML/ANE). It is released under MIT.

What it does. Fast, private push-to-talk dictation for Apple Silicon Macs — local Parakeet TDT v3, no cloud or telemetry.

Presspeech is listed with current version 0.3.9.

Full profile and verified facts for Presspeech →

How to choose

• Free of charge. Aiko, BeMyEars and Google AI Edge Eloquent — no payment is required.
• No licence fee, source available. Azex Speech, buzz, LocalScribe, LokalBot, OpenDictation and Presspeech — the source can be read and, under the terms of each licence, reused.

Open source: Azex Speech, buzz, LocalScribe, LokalBot, OpenDictation and Presspeech (6 of 9).

Common mistakes

• Installing a dictation tool to transcribe recordings, or a transcription tool to dictate. They look like the same category and share almost no workflow.
• Reading “on-device” as “never uses the network”. Check what the app sends, because the two statements are not the same one.
• Judging a transcription engine by word error rate alone. Punctuation, speaker separation and paragraphing are what make a transcript readable, and they are not in that number.
• Pointing a dictation tool at a meeting with several speakers. Dictation assumes one voice, and it will attribute every sentence to whoever it is tracking.

How this guide was put together

Each app’s facts come from its own record: the developer’s own pages and repository, the App Store listing where one exists, and the documentation, each with a source grade and an access date. Where a fact rests on the developer’s own statement rather than on something we verified independently, the sentence says so. Where our data has no value for something, the guide says “not published” instead of guessing.

A price appears here only where the record carries one from a stated source; the guide never estimates one, and it records the pricing model alongside so the two can be read together. Prices change and vary by region, so each app links to its own page for the current detail.

The facts and the recommendations are kept apart on purpose. Everything in the app entries, the comparison table and the answers above is composed from the records. The short answer, the criteria and the picks are our editorial judgment: they reflect what we would choose and why, and they are not paid for. Where we make a recommendation we name the condition it applies under, because we do not call anything “the best” without explaining the criteria.