Wispr Flow logo

Alternatives to Wispr Flow

Explore the best alternatives to Wispr Flow. Compare features, pricing, integrations and see which tool fits your needs.

Last verified: September 28, 2026

Why people switch away from Wispr Flow:

  • Subscription cost and word caps: Pro costs $12/month, while the free plan is capped at 2,000 words/week

  • Cloud dependency: Audio is processed in the cloud, creating privacy concerns for some users

  • High CPU and RAM usage

  • Weaker mobile and Windows experience: Setup friction and occasional bugs compared with the Mac experience

TL;DR:

The best overall alternative to Wispr Flow is Superwhisper, especially if you want more control over your dictation with local models, offline support, and custom modes. For a free alternative, Typeless offers a generous 8,000 words per week, while FluidVoice is a great choice if you want something free, open source, and fully local.

If you primarily use voice for coding, Aqua Voice stands out for its low latency and ability to use screen context to understand code and technical terms. And if you want dictation and meeting notes in one app, Monologue combines system-wide dictation, voice notes, bot-free meeting transcription, and MCP support.

#1
Superwhisper logo

Superwhisper

AI voice-to-text for Mac, Windows, iOS, and Android that uses local-first processing and programmable modes to give users control over models, tone, structure, and formatting.

Superwhisper app screenshot

Features

  • Voice-to-text: Works in any app including Slack, Cursor, Notion, email, and terminals

  • 100+ languages: Supports languages and dialects, with translate-to-English

  • Custom modes: Define tone, structure, and formatting for tasks such as messages, emails, code, and custom workflows

  • Model choice: Choose between cloud and local models, with full offline support

  • BYO AI API keys: Available on Pro

  • Super Mode: Uses AI and adapts to your screen for smarter results

  • Meeting transcription: Records and transcribes meetings, plus audio and video files

  • Custom vocabulary: Supports names, abbreviations, and technical terms

  • Dictation controls: Mouse-start dictation, clipboard auto-paste, and push-to-talk

Pricing

  • Free: $0, including voice-to-text in any app, meeting recording and transcription, 100+ languages, unlimited Whisper models, custom prompt control, and a 3,000-word Pro trial

  • Pro: $8.49/mo, with 40% off for students

  • Lifetime: One-time purchase, described as their “top choice” tier

  • Enterprise: Custom, including SOC 2 Type II, centralized billing and authentication, and model access control

  • Refund: 30-day refund on all plans

Why it's better than Wispr Flow

  • Local processing: Runs models on-device for privacy and offline use

  • One-time purchase: Lifetime option instead of a $12/month subscription

  • Programmable modes: Gives users more control over output structure and formatting

What you might lose if you switch

  • Less complete meeting functionality: Superwhisper provides meeting recording and basic notes, but lacks Wispr's speaker identification, action-item summaries, and MCP pipeline

  • No Snippets: No speak-a-shortcut text expansion

  • Less team and enterprise functionality: No shared dictionary and snippets, HIPAA BAA, SSO, SCIM, or MDM

#2
Aqua Voice logo

Aqua Voice

AI dictation app that inserts text in sub-second time and uses screen context to improve accuracy with code, variable names, and technical terms.

Aqua Voice app screenshot

Features

  • Streaming dictation: Sub-second insertion at approximately 450ms to 1s, with claimed speeds of 230 WPM compared with 40 WPM typing

  • Deep Context: Privately reads your screen to improve recognition of code, variables, and proper nouns

  • Avalon model: Proprietary model with 97.3% accuracy on the AISpeak benchmark

  • Custom dictionary: Supports names, brands, technical terms, and custom instructions

  • Per-app writing styles: Different rules for apps, such as lowercase in Slack or structured output in documents

  • 49 languages

  • Real-time refinement: Handles grammar and formatting while you speak

  • Transcript privacy: Private transcript history and Privacy Mode, where nothing is stored on servers

  • Developer support: Works in terminals, IDEs, Codex, Cursor, Claude, and ChatGPT

  • Platforms: Mac and Windows

Pricing

  • Starter: Free, with 1,000 free words and the Avalon transcription model

  • Pro: $8/mo, with 20% off on annual billing

  • Max: $24/mo, including everything in Pro plus AI power-user features

Why it's better than Wispr Flow

  • Lower latency: Streaming sub-second insertion compared with Wispr's stop-and-insert behavior

  • Screen-context accuracy: Improves recognition of code and technical terms, including Tailwind class names

  • Lower Pro price: $8/month compared with Wispr's $12/month

What you might lose if you switch

  • No mobile: Only Mac and Windows

  • No Meeting Notetaker: No meeting-notes product

  • Fewer languages: 49 languages compared with Wispr's 100+

  • No team and enterprise features: No SSO, admin controls, or HIPAA BAA

  • No Snippets: No text expansion

#3
Willow Voice logo

Willow Voice

Y Combinator-backed AI dictation that turns rambling speech into clean, send-ready text, with offline processing, privacy controls, and a bundled AI writing layer called Scribe.

Willow Voice app screenshot

Features

  • System-wide dictation: Works in Slack, Gmail, Cursor, Notion, iMessage, and other apps

  • Frontier Pro: Fast and accurate speech-to-text with approximately 200ms text output

  • Willow Scribe: AI writing from intent

  • Smart memory: Learns your writing style and dictionary automatically

  • Smart correction and formatting: Includes shortcuts for saved phrases

  • 100+ languages: Includes whisper mode for noise, whispering, and fast speakers

  • Offline mode: Runs on-device

  • Privacy and compliance: Privacy Mode, SOC 2 Type II, HIPAA, and enforced zero data retention on Business+

Pricing

  • Basic: Free, with unlimited use of the weaker Frontier Mini model and 20 Scribe uses/week

  • Pro: $15/mo monthly or $12/mo annually, saving 20%

  • Business: $35/mo monthly or $28/mo annually

  • Enterprise: Custom

  • Discounts: Available for students, educators, nonprofits, and veterans

Why it's better than Wispr Flow

  • Privacy: Enforced zero data retention, SOC 2, HIPAA, and offline mode

  • AI writing layer: Scribe adds AI writing beyond dictation

  • Hands-free activation: “Hey Willow” wake-word invocation

What you might lose if you switch

  • No Android: Android is still coming soon

  • No Meeting Notetaker: No meeting-notes product

  • Weaker free-tier model: Basic uses the weaker Frontier Mini model

  • No Snippets: No text expansion

  • Polish: Some users report occasional latency spikes and language-handling quirks

#4
Typeless logo

Typeless

AI dictation app that learns your personal writing style and automatically edits, formats, and adapts your speech so the output reads like you typed it.

Typeless app screenshot

Features

  • Filler removal: Removes “um,” “uh,” “you know,” and repetition

  • Self-correction: Automatically edits mid-sentence corrections

  • Auto-formatting: Converts speech into lists, steps, and key points

  • Personalized style and tone: Learns your writing habits over time

  • Personal dictionary: Automatically and manually maintained

  • 100+ languages: Includes automatic detection and seamless code-mixing

  • Per-app tones: Uses different tones for different apps

  • Translation: Translates as you speak

  • Ask anything: Lets you speak to edit selected text, make it shorter or longer, change tone, summarize, explain, translate, search, or brainstorm

  • Whisper mode

  • Privacy: Zero cloud retention, no training on your data, and on-device history

  • Platforms: macOS, Windows, iOS, Android

Pricing

  • Free: $0, with 8,000 words/week and standard accuracy

  • Pro: $12/mo billed yearly or $30/mo billed monthly

  • Enterprise: Custom, including SSO, SCIM, HIPAA, BAA, and volume discounts

  • Compliance: GDPR, ISO 27001, HIPAA-ready

Why it's better than Wispr Flow

  • Larger free tier: 8,000 words/week compared with Wispr's 2,000

  • Deeper personalization: Learns your writing style over time

What you might lose if you switch

  • No Meeting Notetaker: Pure dictation and editing with no meeting transcription

  • No Snippets: No text expansion

  • No MCP meeting-notes pipeline

  • Standard free-tier accuracy: Free uses standard accuracy while Pro uses enhanced accuracy

#5
FluidVoice logo

FluidVoice

Free, open-source macOS dictation that turns rough speech into polished text in any app while processing everything on-device.

FluidVoice app screenshot

Features

  • Real-time dictation: Approximately 3.7x faster than typing with less than 100ms perceived latency

  • System-wide input: One shortcut works in email, documents, chats, terminals, and code editors

  • Offline: 0 bytes leave your Mac

  • Fluid Intelligence and Fluid-1: Custom-trained local model of approximately 3.5GB for filler removal, casing, structure, and tone

  • Dictation modes: Write Mode, Command Mode, and Direct Dictation

  • Languages: 40+ languages with Nemotron Speech 3.5 and up to 99 with Whisper

  • Apple Silicon optimization: CoreML and Metal support, with Intel supported through Whisper

  • AI post-processing: Supports Fluid Intelligence, OpenAI, Groq, or custom providers

  • Adaptive app tone: Different tones for Slack, Mail, GitHub, and Notes

  • Models: Nemotron Speech 3.5, Parakeet Flash/TDT v3/v2, Cohere Transcribe, Apple Speech, and Whisper

  • Platforms: macOS 15.0+, with iOS and Windows shown as coming

Pricing

  • Free forever: $0 with no tiers

  • License: GPLv3 open source

  • Installation: brew install --cask fluidvoice

Why it's better than Wispr Flow

  • Free and open source: Free forever instead of a $12/month subscription

  • Fully local: No cloud processing and works offline

  • Fluid-1 polish: Their site and Reddit comparison tests describe the custom-trained model as “just as good as Wispr Flow” for polish

What you might lose if you switch

  • macOS only: No Windows, iOS, or Android

  • No Meeting Notetaker: No meeting-notes product

  • No enterprise features: No team tier, SSO, or HIPAA

  • No Snippets: No text expansion

  • Large optional download: Fluid-1 is approximately 3.5GB

  • Intel limitation: Intel Macs only get Whisper and do not get Fluid-1 or ANE speed

#6
VoiceInk logo

VoiceInk

Open-source macOS dictation that runs locally across apps and uses a one-time purchase model.

VoiceInk app screenshot

Features

  • Local AI models: Parakeet and WhisperKit, with offline support and near-perfect accuracy

  • System-wide dictation: Works in any app

  • Faster writing: Emails and AI chat claimed to be 5x faster

  • Custom words: Dictionary for names, product names, and terms

  • Smart Replace: Phrase expansion

  • Enhancement: Filler removal, note structuring, inbox writing, rewriting selected text, and journal entries

  • Markdown journal: Integrated Markdown journal support

Pricing

  • Solo: $25 one-time for 1 Mac

  • Personal: $39 one-time for 2 Macs

  • Extended: $49 one-time for 3 Macs

  • Regular pricing: $29/$49/$69, currently 50% off

  • Updates: Lifetime updates

  • Refund: 14-day money-back

  • Download: Free to download

  • Requirements: Apple Silicon only, macOS 14.4+

Why it's better than Wispr Flow

  • One-time pricing: $25 to $49 instead of $12/month

  • Local and open source: Voice stays on the Mac and works offline

What you might lose if you switch

  • No Meeting Notetaker: No meeting-notes product

  • No streaming: Waits until you finish speaking, then takes 2 to 4 seconds before text appears

  • Limited platforms: macOS and iOS only, with no Windows or Android

  • Apple Silicon only

  • No team or enterprise tier

#7
Spokenly logo

Spokenly

AI-powered dictation for Mac, Windows, Linux, and iPhone that lets you speak at the cursor using local models or your own API keys.

Spokenly app screenshot

Features

  • 100+ languages: Includes mid-sentence mixing and automatic language detection

  • Local-only mode: 0 network requests

  • AI Instructions: Grammar fixes, filler removal, and app-specific formatting

  • Live streaming: Words appear as you speak

  • History: Searchable and exportable history with audio replay

  • Meeting recording: Captures microphone and system audio without a bot and transcribes later

Pricing

  • Local Models: $0 forever, with Whisper and Parakeet, offline use, unlimited dictation, and no account

  • Your API Keys: $0 forever, including local features plus OpenAI, Deepgram, and Groq real-time models

  • Pro: $8.33/mo, billed at $99.99/year, with cloud models, sync, and AI cleanup

  • Refund: 14-day refund

Why it's better than Wispr Flow

  • Free local models: Unlimited use compared with Wispr's 2,000-word cap and $12/month price

  • BYO API keys: Supports multiple providers with their own free tiers

  • Linux support: Cross-platform support includes Linux, which Wispr does not support

What you might lose if you switch

  • No Meeting Notetaker: Meeting recording is available, but there are no summaries, speaker identification, or decision extraction

  • Cloud accuracy: Best cloud accuracy requires Pro or your own API keys

  • No Android: Android is not yet available

  • No Snippets: No text expansion

#8
Voice Gecko logo

Voice Gecko

Free, open-source desktop dictation that runs fully locally and turns speech into clipboard-ready text for emails, coding, AI prompts, and brain dumps.

Voice Gecko app screenshot

Features

  • Local speech-to-text: Transcripts and dictionary stored in local SQLite and never uploaded

  • Desktop app support: Works with Slack, Jira, Discord, GitHub, VS Code, and Telegram

  • Fast dictation: Approximately 5x faster than typing, at 200 vs 40 WPM

  • GeckoBar: Persistent bar for triggering dictation and viewing status

  • Clipboard output: Instant paste

  • AI prompt dictation: Speak prompts to an LLM

  • Custom dictionary: Supports names and jargon

  • Custom shortcuts

  • Searchable history

  • Privacy mode

Pricing

  • Free: $0 forever, MIT licensed, full product, unlimited dictations, no weekly limits, and no account required

  • Support: Optional $5.99/mo to fund development, with the same product

Why it's better than Wispr Flow

  • Free and unlimited: No word cap or subscription

  • Open source: MIT licensed

  • Fully local: Voice is never uploaded

  • Lightweight workflow: Clipboard-first capture

What you might lose if you switch

  • Windows only: No macOS, iOS, or Android

  • No AI rewrite layer: Focuses on transcription rather than Wispr's auto-editing and formatting

  • No Meeting Notetaker: No meeting transcription workflow

  • Fewer languages: Local speech recognition is English-centric compared with Wispr's 100+ languages

#9
TalkTastic logo

TalkTastic

Context-aware macOS voice dictation that analyzes the app you're using to match its tone, style, vocabulary, and context.

TalkTastic app screenshot

Features

  • Context-aware dictation: Analyzes a snapshot of the current app to match tone, style, and substance

  • Smart rewrites: Keeps the user's style and voice

  • AI transcripts: Combines Apple Dictation, on-device Whisper, and multimodal LLMs

  • Screen context: Helps recognize names and easily confused words

  • Privacy controls: Snapshot only on command, optional auto-delete, and no training on user data

Pricing

  • Free: 100% free while in beta

  • Requirements: macOS 13.1+

  • Paid tiers: None yet

Why it's better than Wispr Flow

  • Free: Free while in beta

  • Screen context: Uses the current app to match tone and recognize proper nouns

What you might lose if you switch

  • macOS only: No Windows, iOS, or Android

  • No Meeting Notetaker: No meeting-notes product

  • Limited language coverage: English-centric with no advertised multilingual coverage

#10
MacWhisper logo

MacWhisper

Mac transcription and dictation software for lectures, videos, podcasts, files, and online meetings, with local transcription and a one-time Pro license.

MacWhisper app screenshot

Features

  • File transcription: Drag and drop mp3, wav, m4a, ogg, opus, mov, and mp4 files

  • Meeting recording: Automatically records Zoom, Teams, Webex, Skype, Discord, and Chime, with automatic start and end detection and no bot

  • System-wide dictation: Pro includes high-quality dictation, AI prompts, app-specific prompts, and grammar

  • Podcast transcription: Creates per-speaker files

  • YouTube transcription

  • App Audio capture

  • Local models: Privacy and offline support

  • Speaker recognition

  • Filler cleanup

  • 100+ languages: Includes translation

  • Exports: .srt, .vtt, .txt, .md, .pdf, .html, and .docx

  • Search and editor

  • Batch transcription

  • Watched Folders

  • AI services: Supports BYO API keys for OpenAI, Anthropic, Groq, xAI, Azure, Deepseek, Google, and OpenRouter

  • Local AI: Ollama and LM Studio

  • AI tools: Summarization, chat, and custom prompts

  • Workflow automation: Auto-upload to Notion, Zapier, Obsidian, n8n, Make.com, or a webhook

  • CLI: Command-line control

Pricing

  • Free: €0, free forever

  • Pro: €64 one-time, with lifetime updates

  • Purchase: Available through Gumroad or the App Store

Why it's better than Wispr Flow

  • One-time pricing: €64 lifetime license instead of a $12/month subscription

  • On-device transcription: Local Whisper provides privacy and offline processing for sensitive files

  • Deep transcription workflow: Supports subtitles, batch processing, meetings, and watched folders

What you might lose if you switch

  • Dictation is secondary: The product is primarily a file transcription tool rather than a live dictation tool

  • Less live-dictation polish: Does not match Wispr's real-time auto-editing and rewriting

  • Limited platforms: macOS and iOS only, with no Windows or Android

  • Free-tier limitations: No speaker recognition, batch transcription, meetings, AI services, or high-quality dictation without Pro

#11
Hex logo

Hex

A free, open-source voice dictation app for Mac, Windows and Linux where you hold a shortcut, speak, and Hex turns your speech into text on your own computer with models such as Parakeet and Whisper, then pastes it into whatever app you're using.

Hex app screenshot

Features

  • Hold-to-talk dictation in any app: Hold the shortcut, speak and release, and the text is pasted where your cursor is. Double-tap to lock recording, or switch to Toggle mode, where you press once to start and again to finish.

  • Transcription on your computer: Speech is turned into text locally, using your graphics chip through Metal on macOS or Vulkan on Linux and x64 Windows. Only the first model download from Hugging Face needs the internet.

  • Model picker by language: About 14 catalog models, including Parakeet Unified English, Parakeet v2 and v3, Whisper large-v3-turbo, Qwen3-ASR 0.6B, SenseVoice Small, Cohere Transcribe, GigaAM v3 for Russian and an ivrit.ai Whisper for Hebrew. The picker shows Recommended, Fastest and Most accurate options for your language, plus each model's accuracy.

  • Bring your own model: Import a speech model saved as a GGUF file, or reuse models another app already downloaded to the Hugging Face cache on your computer.

  • Vocabulary list: Add names and terms, and Hex steers transcription toward them and fixes common mishearings. Since version 3.3.0, each Mode can have its own list.

  • Modes for each app or website: Modes run corrections, text replacements, optional AI rewriting through OpenCode and your own TypeScript transformations on the finished text, with different settings per app or site.

  • History and Paste Last: Keeps past dictations for seven days by default and re-pastes the last one with a shortcut (Ctrl+Alt+V on Windows). Audio is saved only if you turn on Labs › Keep recordings.

  • Flexible shortcuts: Use a modifier key on its own, both Option keys together, a left or right modifier, a key combination or a mouse button.

  • Meeting recorder (macOS beta): Opt-in call recording with a live transcript, calendar reminders and notes summarized by Apple Intelligence or, if you turn it on, OpenCode. It's off by default.

  • Command line and agent access: The hex command controls the running app (hex status, hex dictation, hex settings) and reads History and meetings. hex mcp gives AI agents read-only access to your meetings.

  • Low memory use: The 3.0.0 release notes say local models now use about 160 MB of memory, down from about 900 MB.

  • Skips silence: Recordings with no speech in them are dropped instead of transcribed.

Pricing

Hex is free. The only possible cost comes from the optional AI rewriting, which runs through your own OpenCode setup and the model provider you connect to it.

  • Free: Open source under the MIT licence, with free downloads for every platform. The repo and site list no paid plan, account or usage limit.

  • OpenCode rewriting: Optional. It uses your own OpenCode install and whichever model provider you set up there, so any cost depends on that provider. Hex doesn't publish a figure.

Why it's better than Wispr Flow

  • Free and open source: Hex is MIT licensed with no word cap. Wispr Flow's free plan stops at 2,000 words a week, and Pro costs $12 to $15 a month.

  • On-device and offline: After the first model download, transcription runs on your computer. Wispr Flow sends every dictation to the cloud.

  • Fast and light: Users say long dictations come back in about a second with Parakeet, and the app uses about 160 MB of memory.

  • Linux support: It runs on Mac, Windows and Linux, and Wispr Flow has no Linux app.

  • Model choice: You can pick from about 14 speech models by language or load your own GGUF file.

What you might lose if you switch

  • No mobile apps: Hex has no iOS or Android release, and Wispr Flow runs on both.

  • Less auto-polish: Wispr cleans up your text out of the box, but in Hex AI rewriting needs your own OpenCode setup.

  • No Notetaker: Meeting recording is a Mac-only beta without Wispr's speaker ID or summaries.

  • Rough edges: The Windows and Linux builds are betas, and the Mac app needs Apple silicon and macOS 15.

  • No team or compliance tier: There's no shared dictionary, SSO, SOC 2 or HIPAA.

#12
shepetal logo

shepetal

A Mac dictation app that turns your speech into text on the Mac itself with Parakeet and Whisper models, so you press one key, speak in any of 100 languages and your words appear in any app, even with no internet connection.

shepetal app screenshot

Features

  • One-key dictation in any app: Press one key (the site shows the fn key) and speak, and the text is pasted where your cursor is. When an app blocks pasting, the words wait on the clipboard so you can paste them yourself.

  • Speech recognition on your Mac: Parakeet TDT 0.6B v3 handles the 25 European languages and Whisper large-v3-turbo the other 75. shepetal picks one based on the languages you choose, or you set it under Speech model in Settings.

  • Works offline after setup: Needs the internet once to sign in and download the models (about 470 MB, plus 574 MB for non-European languages). After that, dictation runs with no connection.

  • Local text cleanup: Tidies up dictation with Qwen3.5 0.8B, or Qwen3.5 2B on Macs with 16 GB of memory or more, also on the Mac. A "My voice" setting tells it how your text should read.

  • 100 languages with auto-detect: English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Ukrainian and Slovene work right after setup. Japanese, Chinese, Korean, Arabic, Hindi and others need one extra download. The Automatic setting picks up whichever language you speak.

  • Formatting for each app: Lays out dictation as an email in Mail or Gmail, has styles from Formal to Excited for work, email and personal messages, and can set any app to Raw so your words go in unchanged.

  • Custom dictionary: Writes names like LinkedIn with your spelling and capitals, and swaps a word you say for the one you want written.

  • Meeting transcription on the Mac: When a call starts, shepetal offers to record it, captures your microphone and the call audio, and keeps the recording and transcript on the Mac. The changelog lists support for up to five speakers and re-transcribing a meeting.

  • Re-transcribe and history: Re-transcribe past dictations with one press, write numbers as digits, and keep your history in ~/Library/Application Support/Shepetal until you delete it.

  • Usage stats: Home shows your words, words a minute and streak for the week. Insights adds the time saved compared with typing. All of it is counted on the Mac.

  • Checked by Apple and self-updating: Installed from a DMG file, notarized by Apple (Apple's check for Mac apps downloaded outside the App Store), and updates itself once installed.

Pricing

You can try shepetal free for 30 days, then pay either once for a lifetime license or a small monthly fee. Prices are in euros.

  • Free trial: 30 days, no card needed. During the trial the app checks your license at least once every 7 days.

  • Lifetime license: €24 once, with updates included. Works offline with no time limit and on more than one Mac signed in with the same Google account. Full refund within 14 days of purchase.

  • Monthly: €2 a month, cancel anytime. A month that has started isn't refunded.

  • Currency and tax: Prices include sales tax or VAT, and checkout shows the total before you pay. A US dollar price isn't published.

  • Launch promo: The Peerlist launch page offered the code PEERLIST for a $4.99 lifetime license to the first 100 people. shepetal's own site doesn't mention it, and whether any of those spots are left isn't published.

Why it's better than Wispr Flow

  • One-time pricing: A lifetime license is €24 (or €2 a month), against Wispr's $12 to $15 monthly subscription.

  • Works offline: Parakeet and Whisper run on the Mac, while Wispr Flow transcribes everything in the cloud.

  • Local cleanup: A small Qwen model tidies your text on the Mac, so the polish doesn't need a server either.

  • Private meetings: It records calls and keeps both the recording and the transcript on your Mac.

  • No analytics: Recordings and history stay in a local folder, and the app sends no usage data.

What you might lose if you switch

  • Mac only: It needs Apple silicon, and Windows, Linux and iPhone versions are still on a waitlist.

  • No free plan: You pay once the 30-day trial ends, while Wispr keeps a free weekly allowance.

  • Online sign-in: You need a Google account and a daily license check, even though dictation works offline.

  • Weaker mixed-language dictation: The maker says German with English terms isn't reliable yet.

  • Brand new: It launched in October 2026, has no independent reviews yet and offers nothing for teams.

#13
Hold to Talk logo

Hold to Talk

A Mac menu bar dictation app where you hold Option-Space, speak and let go, and your words are pasted into the app you're typing in, using cloud transcription that keeps no audio on its servers.

Hold to Talk app screenshot

Features

  • Hold-to-talk shortcut: Hold Option-Space, speak and release. You can change the shortcut in Settings, and the microphone is only on while you hold the key.

  • Pastes into the app you're using: Text lands in the focused field in Slack, browsers, email, notes, code editors and Terminal, so there's nothing to copy across. A few secure or custom text fields may refuse the paste.

  • Review before sending: The standard shortcut pastes editable text and never submits it. The App Store listing also describes an optional Option-Shift-Space shortcut that pastes and presses Return for when you do want to send.

  • Custom vocabulary: Add names, work terms and odd spellings to a list so they come out right.

  • Fast cloud transcription: Uses Whisper Large v3 on Groq by default, and short dictations often show up in under a second on a fast connection. The privacy policy also lists an optional realtime streaming mode and optional cleanup and "most accurate" modes that use OpenAI.

  • No audio kept on servers: Audio travels over HTTPS, gets transcribed and is discarded. Hold to Talk doesn't store, log or replay audio or transcripts on its servers, and doesn't train models on them.

  • Transcript history on your Mac: Recent transcripts are saved locally, and you can clear the history or turn it off. Recordings that fail to transcribe stay on your Mac so you can recover them.

  • Menu bar popover: Shows your shortcut, weekly usage and recent transcripts, and opens settings for the shortcut, vocabulary, paste behavior and privacy.

  • Two permissions only: It asks for Microphone access and a limited paste permission that macOS lists under Accessibility, which it uses only to press Command-V. It doesn't use Input Monitoring or read other apps' screens.

  • Switch mics mid-recording: Version 1.2.75 (November 2026) lets you move between AirPods and your Mac's microphone during a recording and warns you if any audio was missed.

  • iPhone and iPad voice keyboard: The separate iOS app lets you dictate inside the app and copy the text, or turn on a voice keyboard for Messages, email, notes and other apps. It has its own vocabulary list.

Pricing

  • Free: $0, with about 2,000 words a week on the Mac. You get the same hold, speak and paste workflow in most standard Mac text fields.

  • Pro: $10 a month for 30 hours of transcription each month. You buy it inside the app through Apple and manage or cancel it in your Apple Account settings. An annual price for the Mac isn't published.

  • iPhone and iPad: A free download that includes 20 free dictations. Pro covers up to 30 hours a month, billed monthly or yearly from inside the app; the website doesn't list the iOS prices.

  • Older web subscriptions: Plans bought earlier on holdtotalk.com through Stripe are still managed on the website. All new purchases go through the App Store.

  • Team or business plan: Not published.

Why it's better than Wispr Flow

  • Lower Pro price: Pro is $10 a month for 30 hours of transcription, while Wispr Flow Pro costs $12 to $15 a month.

  • No audio kept: Audio is discarded as soon as the text comes back and is never used to train models, and your transcript history stays on your Mac.

  • Your words as spoken: It pastes a plain transcript with no AI rewriting, which suits prompts, code comments and Terminal commands where you want exact wording.

  • Nothing sends by itself: The standard shortcut only pastes, so you check the text before a message, prompt or command goes anywhere.

  • Fewer permissions: It asks only for the microphone and a limited paste permission, with no Input Monitoring and no reading other apps' screens.

What you might lose if you switch

  • No auto-polish: Wispr cleans up and formats your text, and Hold to Talk has no AI cleanup layer.

  • No Windows or Android: It runs on Mac, plus a separate iPhone and iPad app whose prices aren't published on the website.

  • No Notetaker: The maker says it isn't a meeting recorder, while Wispr includes its Notetaker even on the free plan.

  • No team or compliance tier: There's no published team plan, shared dictionary, SSO, SOC 2 or HIPAA option.

  • New and unreviewed: The Mac app launched in August 2026, both App Store listings have no ratings yet, and the supported languages aren't published.

#14
HoldTalk logo

HoldTalk

A Mac dictation app where you hold the fn key, speak and let go, and the text appears at your cursor in any app, transcribed on your Mac by Apple's speech engine and sold for a one-time price.

HoldTalk app screenshot

Features

  • Hold-to-talk in any app: Click in a text field, hold fn, speak and release. The text is pasted at the cursor anywhere you can type, including Gmail, Outlook, Word, Slack and WhatsApp.

  • Hands-free mode: For long dictation, tap fn once and keep talking without holding the key down.

  • Automatic punctuation: Talk normally without saying "comma" or "period". Punctuation, times and amounts are written for you.

  • Quick insertion: The site says text appears about 0.3 seconds after your last sentence.

  • Transcription on your Mac: Apple's SpeechAnalyzer engine from macOS 26 does the work locally. Audio is deleted after transcription and is never recorded or uploaded, and dictation works offline once a language has been downloaded.

  • English, French and Spanish: HoldTalk dictates in all three, and the app, site and support are available in each.

  • Personal dictionary: Add names, brands and trade terms you use often. After each dictation, HoldTalk spells them the way you typed them and fixes words it misheard.

  • Searchable history: Your last 1,000 texts stay on the Mac, each with the name of the app it went into, so you can find one that was pasted in the wrong place. One click clears the list.

  • Clipboard and sound handled: Your clipboard goes back to what it held before each paste, and the Mac's sound is muted while you dictate.

  • Custom shortcut: fn is the default. If another app already uses fn, you can choose a different key in HoldTalk.

  • No account: The free version works as soon as it's installed. Pro is unlocked with a key sent by email.

  • Text cleanup (coming): Listed as an upcoming Pro feature that existing Pro buyers will get at no extra cost. It hasn't shipped yet.

Pricing

  • Free: 7 days of unlimited dictation from the first launch, with no card or account. After that, up to 2,000 words a week, which the site puts at about 15 to 20 minutes of dictation. All three languages, hands-free mode and recent texts are included. The count resets every Monday, a dictation in progress finishes even if it crosses the limit, and words from the trial week don't count.

  • HoldTalk Pro: 19 euros once until November 15, 2026, then 29 euros. Unlimited dictation, one key for up to 2 Macs, free updates throughout version 1, and every current and future version 1 Pro feature, including text cleanup when it arrives. There's no subscription. A future major version may be offered to key holders at a reduced price.

  • Refund: Full refund within 14 days of purchase, with no reason needed.

  • Teams: No team plan is published. The terms ask teams to email support.

  • Payment: Lemon Squeezy sells the licenses and handles payment, invoices and VAT. Prices are listed in euros only.

Why it's better than Wispr Flow

  • Pay once: Pro is a one-time €19 until November 15, 2026 (€29 after) and covers 2 Macs, against Wispr's $12 to $15 monthly subscription.

  • On-device and offline: Apple's speech engine transcribes on the Mac and the audio is deleted afterwards, while Wispr Flow sends dictation to the cloud.

  • No account: The free version works as soon as it's installed, and Pro is just a key you paste in.

  • Same free allowance, plus a trial: You get 7 days of unlimited dictation, then 2,000 words a week, the same weekly cap as Wispr's free desktop plan.

  • Handy history: It keeps your last 1,000 texts with the app each one went into, and it restores your clipboard after every paste.

What you might lose if you switch

  • Newer Macs only: It needs Apple silicon and macOS 26, with no Windows, iPhone or Android version.

  • Three languages: It dictates in English, French and Spanish, while Wispr covers 100+.

  • No AI cleanup yet: Text cleanup is listed as a coming Pro feature with no release date.

  • No Notetaker or team plan: There's no meeting recording, no published team pricing and no SOC 2 or HIPAA option.

  • Brand new: It launched in early November 2026, has no independent reviews yet and is priced only in euros.

Check out what's new today