Vendor page

open-wispr: Free Voice Dictation for macOS (100% Local, No Cloud)

open-wispr (github.com/human37/open-wispr), 2026

Read the original

Read on October 20, 2026

What it is: The product site of open-wispr, a free, MIT-licensed dictation app for macOS by the human37 project. It describes on-device transcription with Whisper, the audio and network behavior, the features and a comparison table with other dictation apps.
How much weight it carries: The maker's own page, so it states what the maker claims. It is a product page with no independent testing. The comparison table is the maker's own and covers competitors selectively. Only the maker's statements are used here.

Facts taken from this source

open-wispr's home page says it supports 99 languages through Whisper's multilingual models.
“Supports 99 languages via Whisper's multilingual models.” (Features, Multilingual)
open-wispr's own comparison table marks 'AI features' as No for open-wispr.
“open-wispr VoiceInk Wispr Flow Superwhisper MacWhisper Apple Dictation ... AI features No AI assistant AI rewriting AI formatting No No” (Compare table)
open-wispr's page says transcribed text is typed at the cursor in any app.
“Text appears at your cursor in any app. Editors, browsers, terminals, Slack, email.” (How it works, step 03)
open-wispr's FAQ says no audio or text leaves the computer.
“Does any audio or text leave my computer? No. Everything runs 100% on-device.” (FAQ)
open-wispr's own comparison table lists its price as free forever.
“open-wispr VoiceInk Wispr Flow Superwhisper MacWhisper Apple Dictation Price Free forever $39.99 $15/mo $8.49/mo $69 lifetime” (Why open-wispr? comparison table)
open-wispr's FAQ says it is free and open source under the MIT license, with no subscriptions, usage limits or premium tier.
“Yes, completely free and open source under the MIT license. No subscriptions, no usage limits, no premium tier.” (FAQ)
open-wispr keeps no past recordings by default, and you can set its maxRecordings option to keep 1 to 100 recent recordings.
“0 = nothing stored (default). Set 1-100 to keep that many recent recordings.” (Configuration options table, maxRecordings)
open-wispr's page says its voice to text is entirely on-device.
“Voice to text, entirely on-device. Hold a key, speak, release. Your words appear at the cursor.” (Hero section)
open-wispr says audio goes to a temporary file, is transcribed on the computer, and is then deleted immediately.
“Audio goes to a temp file, transcribed locally by Whisper, then deleted immediately.” (FAQ, 'Does any audio or text leave my computer?')
open-wispr's FAQ says no audio or text leaves the computer, so no third-party AI receives its text.
“Does any audio or text leave my computer? No. Everything runs 100% on-device.” (FAQ)
open-wispr's page says dictation makes zero network requests and involves no cloud.
“Audio is recorded to a temp file, transcribed locally, and deleted. Zero network requests. No accounts, no telemetry, no cloud.” (Private feature)

Vendor page

open-wispr: Free Voice Dictation for macOS (100% Local, No Cloud)

open-wispr (github.com/human37/open-wispr), 2026

Read the original

Read on October 20, 2026

What it is: The product site of open-wispr, a free, MIT-licensed dictation app for macOS by the human37 project. It describes on-device transcription with Whisper, the audio and network behavior, the features and a comparison table with other dictation apps.
How much weight it carries: The maker's own page, so it states what the maker claims. It is a product page with no independent testing. The comparison table is the maker's own and covers competitors selectively. Only the maker's statements are used here.

Facts taken from this source

open-wispr's home page says it supports 99 languages through Whisper's multilingual models.
“Supports 99 languages via Whisper's multilingual models.” (Features, Multilingual)
open-wispr's own comparison table marks 'AI features' as No for open-wispr.
“open-wispr VoiceInk Wispr Flow Superwhisper MacWhisper Apple Dictation ... AI features No AI assistant AI rewriting AI formatting No No” (Compare table)
open-wispr's page says transcribed text is typed at the cursor in any app.
“Text appears at your cursor in any app. Editors, browsers, terminals, Slack, email.” (How it works, step 03)
open-wispr's FAQ says no audio or text leaves the computer.
“Does any audio or text leave my computer? No. Everything runs 100% on-device.” (FAQ)
open-wispr's own comparison table lists its price as free forever.
“open-wispr VoiceInk Wispr Flow Superwhisper MacWhisper Apple Dictation Price Free forever $39.99 $15/mo $8.49/mo $69 lifetime” (Why open-wispr? comparison table)
open-wispr's FAQ says it is free and open source under the MIT license, with no subscriptions, usage limits or premium tier.
“Yes, completely free and open source under the MIT license. No subscriptions, no usage limits, no premium tier.” (FAQ)
open-wispr keeps no past recordings by default, and you can set its maxRecordings option to keep 1 to 100 recent recordings.
“0 = nothing stored (default). Set 1-100 to keep that many recent recordings.” (Configuration options table, maxRecordings)
open-wispr's page says its voice to text is entirely on-device.
“Voice to text, entirely on-device. Hold a key, speak, release. Your words appear at the cursor.” (Hero section)
open-wispr says audio goes to a temporary file, is transcribed on the computer, and is then deleted immediately.
“Audio goes to a temp file, transcribed locally by Whisper, then deleted immediately.” (FAQ, 'Does any audio or text leave my computer?')
open-wispr's FAQ says no audio or text leaves the computer, so no third-party AI receives its text.
“Does any audio or text leave my computer? No. Everything runs 100% on-device.” (FAQ)
open-wispr's page says dictation makes zero network requests and involves no cloud.
“Audio is recorded to a temp file, transcribed locally, and deleted. Zero network requests. No accounts, no telemetry, no cloud.” (Private feature)