OpenWhispr

OpenWhispr is an open-source voice-to-text assistant for dictating in apps and turning audio into transcripts and notes. It supports more than 100 languages with automatic language detection, and dictation works in apps that accept text, including ChatGPT, Claude, Cursor, Google Docs, Gmail, Slack, and Teams. Its meeting note taker can capture computer audio without joining calls as a bot, then create structured notes with decisions and action items. It works with Zoom, Google Meet, Microsoft Teams, Slack, Discord, Webex, and other apps that play computer audio. Audio can be processed locally on the user's device, or through OpenWhispr Cloud, which says it discards audio without retaining it. The company says it does not use user content to train AI models. The Free plan includes 2,000 transcription words per week and five hours of meeting recordings per month. Pro costs $6.67/user/mo billed annually and includes unlimited cloud transcription and 20 hours of meeting recordings per month. OpenWhispr is available for iOS, Linux, macOS, and Windows.

OpenWhispr combines in-app dictation with meeting transcription and notes, and offers both local and cloud processing. Check the free plan’s word and recording limits, or the annual Pro price, against your needs.

OpenWhispr may suit people who dictate into text-entry apps or want meeting notes from computer audio. Its local processing option is for users who want audio kept on their device.

OpenWhispr is an open-source voice-to-text assistant that combines in-app dictation, audio transcription and meeting notes. It is best suited to people who move between writing, calls and recorded audio and want a choice between on-device and managed cloud processing. The free plan is a credible trial, while frequent transcription or meeting capture requires careful attention to its quotas and annual billing.

Overview

OpenWhispr puts dictation directly into applications that accept text, including ChatGPT, Claude, Cursor, Google Docs, Gmail, Slack and Teams. Its meeting note taker captures computer audio without joining a call as a bot, then can create structured notes containing decisions and action items. That makes it a practical single workspace for both spontaneous speech and formal meetings, rather than a tool limited to one recording workflow.

The project is open source under the MIT license. Local speech models keep audio on the device, while OpenWhispr Cloud handles audio through a managed real-time API and says it discards the audio instead of retaining it. OpenWhispr also says user audio, transcriptions and other content are not used to train AI models.

Key features

Dictation and transcription

Support for more than 100 languages with automatic language detection is a strong fit for multilingual users or mixed-language work. Uploads accept MP3, WAV, M4A, WebM, OGG, FLAC, AAC and formats supported by FFmpeg. A transcript editor, speaker labels, caption export and searchable library make the product useful after the words have been captured, not just while dictating.

Meeting notes

The note taker works with Zoom, Google Meet, Microsoft Teams, Slack, Discord, Webex and other applications that play computer audio. It needs no plugins or integrations, and its bot-free capture avoids placing a separate participant in a call. The trade-off is capacity: the Free plan allows five hours of meeting recordings per month, while Pro raises that to 20 hours and Business removes the recording limit.

Privacy and integrations

Cloud data is encrypted in transit with TLS and at rest, and diagnostic reporting is off by default. The homepage states HIPAA compliance and independent verification for AICPA SOC 2 and ISO 27001. Pro adds personal API access and MCP integration in apps, useful for users who want to connect transcription to their own workflows rather than stay inside the desktop interface.

Pricing

Free

Free costs $0.00 and is always free. It includes 2,000 transcription words per week, five hours of meeting recordings per month, unlimited local AI models, and unlimited cloud transcription when you supply your own API keys. Community support is the price of entry. This is enough for occasional dictation and testing local models, but the weekly word ceiling can arrive quickly for regular writers.

Pro

Pro costs $6.67 per user per month, billed as $80 per user annually. It removes the cloud-transcription limit, provides 20 meeting-recording hours monthly, and adds Agent mode, device sync, personal API access, MCP integration, and the iPhone and iPad app. Email support replaces community support. Pro is the sensible individual plan for sustained cloud use, provided annual payment suits you.

Business and Enterprise

Business costs $13.33 per user per month, billed as $160 per user annually. It keeps unlimited cloud transcription, makes meeting recordings unlimited, and adds chat over your data and priority support. Enterprise uses custom pricing and adds team and administrative features, including SSO, SAML and SCIM, audit logs, retention controls and dedicated support. Those plans target organizations; they offer little reason for a solo user to pay beyond Pro.

Platforms

OpenWhispr is available on iOS, Linux, macOS and Windows. Pro specifically includes the iPhone and iPad app, while the desktop platform coverage suits users who dictate across operating systems. Meeting capture depends on audio playing through the computer, so it can cover supported call applications without a dedicated integration.

Who it's for

Choose OpenWhispr if you want one tool for dictating into everyday apps, transcribing uploaded files and recording meetings without a bot. Multilingual users benefit from automatic language detection, and privacy-conscious teams can favor local models or the stated cloud data controls. It is less suitable for heavy meeting schedules on Free, or for users unwilling to commit to annual billing for the paid tiers.

Pros and cons

• Pro: Local models keep audio on the device, giving privacy-sensitive users an alternative to cloud processing.
• Pro: Bot-free computer-audio capture works across major meeting apps without plugins or integrations.
• Pro: More than 100 languages, speaker labels, editing, caption export and search cover a broad post-transcription workflow.
• Con: Free transcription is capped at 2,000 words per week and meetings at five hours per month.
• Con: Pro and Business prices are annual commitments, even though they are displayed as monthly rates.
• Con: Advanced API, MCP and administrative controls begin on paid plans, limiting the free tier for automation-heavy work.

Alternatives

For a wider comparison of specialist tools, see Podcast Transcription Software and Meeting Recording Software.

• Happy Scribe is worth considering if unlimited meeting recordings on its free plan and a 45-minute per-recording limit fit your use better.
• Simon Says suits buyers who prefer pay-as-you-go transcription at $15 per hour or $0.25 per minute.
• AudioPod AI is a credits-based option for users whose needs center on mixed audio and media generation.
• Rev is an alternative for English-only users who want 45 free AI transcription and caption minutes monthly.
• Transkriptor offers a 300-minute monthly Lite plan for individuals starting out.
• Trint is a paid transcription alternative with a free trial.
• Castmagic is aimed at paid content workflows, with its Hobby plan offering 30 transcribed hours and three seats.
• AssemblyAI is better suited to developers seeking API access and usage-based billing.

Verdict

OpenWhispr is a strong choice for individuals and small teams that want dictation, bot-free meeting capture and searchable transcription in one open-source product, especially when local processing matters. Choose Pro for regular cloud work and cross-device use; choose Business when unlimited meeting recording is essential. Look elsewhere if you need a nonannual subscription or if the Free quotas cannot cover your routine.