Press a hotkey anywhere on your system, speak naturally, and Telvr transcribes your voice in real-time using Whisper. The finished text is automatically inserted at your cursor position — no copy-paste, no app switching.

Which languages are supported?

Telvr supports 50+ languages via OpenAI's Whisper large-v3 model. Language detection is automatic — just speak in your preferred language and Telvr handles the rest.

Do I need a subscription?

No. Telvr uses a pay-as-you-go model: EUR 3 per month infrastructure fee plus EUR 0.03 per minute of usage. No lock-in, no auto-renewal. You top up your balance and use it at your own pace.

Does Telvr work offline?

Currently, Telvr requires an internet connection for cloud-based transcription via Groq. A Community Edition with local processing using your own API key is planned for the future.

Which apps are supported?

Telvr works system-wide — it inserts text at the cursor position in any application. Email clients, chat apps, code editors, browsers, word processors — if you can type in it, Telvr works there.

All data is transmitted via TLS encryption. Audio recordings are not permanently stored after transcription. Groq processes your audio under a Data Processing Agreement (DPA). We do not sell or share your data.

← Blog2026-02-10

Voice Typing vs Keyboard: A Productivity Analysis

The Speed Myth

The average person types at 40 words per minute. The average person speaks at 130 words per minute. On paper, voice input is 3x faster. In practice, the productivity equation is more nuanced.

Raw speed is only one factor. The total time from thought to finished text includes thinking, input, editing, and formatting. Let us break down each phase.

Input Speed: Voice Wins Clearly

For pure input speed, speaking dominates:

Typing: 40-80 WPM for most professionals. 100+ WPM for fast typists.
Voice: 130-160 WPM for natural speech. Consistent regardless of typing skill.

The gap is especially significant for people who are not fast typists — which is most people outside of software development and writing professions.

Editing Time: It Depends on the Tool

Raw dictation produces text that needs editing. Filler words, missing punctuation, grammar issues, and wrong formatting all require manual cleanup. This editing overhead can erase the speed advantage.

This is where AI enrichment changes the equation:

Without enrichment: You speak 3x faster but spend 2x longer editing. Net productivity gain is modest.
With enrichment (Telvr): The AI removes fillers, fixes grammar, and formats output. Editing drops to minor touch-ups. Net productivity gain is 2-3x.

The quality of the AI processing determines whether voice input is a net productivity win or just a different way to create work.

Cognitive Load: The Hidden Factor

Typing requires translating thoughts into finger movements on a keyboard. Speaking requires translating thoughts into words. For most people, the second translation is more natural and requires less cognitive effort.

This matters most for:

First drafts: Speaking your thoughts flows more naturally than typing them. Ideas come out more complete and coherent.
Long-form content: Writer's block is partly a typing problem. Speaking eliminates the mechanical barrier between thinking and output.
Multilingual work: If you think in one language but type in another, voice input in your thinking language produces better initial output.

However, typing has advantages for:

Structured content: Code, spreadsheets, and formatted documents benefit from the precision of keyboard input.
Quiet environments: Open offices, libraries, and shared spaces make voice input impractical.
Editing and revision: Moving text, selecting words, and making precise changes is faster with keyboard and mouse.

Real-World Scenarios

Email (Voice Wins)

Writing a 200-word email:

Typing: ~5 minutes (including thinking and editing)
Voice with Telvr Email mode: ~2 minutes (speak for 90 seconds + review)

Voice is 2-3x faster because emails are conversational by nature, and the AI handles formatting.

Code Documentation (Voice Wins)

Writing a function description or README section:

Typing: Switching between code and documentation contexts creates friction
Voice: Describe the function naturally while looking at the code. Cleanup mode produces clean documentation.

Slack Messages (Voice Wins)

Quick team updates and responses:

Typing: Even short messages take 30-60 seconds with context switching
Voice: 10-15 seconds including the push-to-talk gesture

Code Writing (Keyboard Wins)

Writing actual code:

Voice: Dictating syntax, brackets, and precise variable names is slower and error-prone
Typing: Direct mapping between thought and characters on screen

Data Entry (Keyboard Wins)

Filling forms, entering numbers:

Voice: Numbers, abbreviations, and field navigation add overhead
Typing: Direct input with Tab key navigation is faster

The Hybrid Approach

The highest productivity comes from combining both input methods:

Voice for content: Emails, messages, documentation, notes, descriptions
Keyboard for structure: Code, formatting, navigation, editing
Context-aware switching: Use the right tool for each micro-task

Telvr's push-to-talk model supports this hybrid approach naturally. Your hands stay on the keyboard. When you need to write text, hold the hotkey and speak. When you need precision, type. There is no mode switching or app changing.

Measuring Your Own Productivity

Try this experiment:

Pick a typical work task (writing an email, documenting a feature, responding to a message)
Time yourself doing it with typing
Time yourself doing the same type of task with voice input
Compare not just speed, but how the output reads

Most people find that voice input produces more natural, complete text on the first attempt — even if the raw speed gain is less than the theoretical 3x.

Conclusion

Voice typing is not about replacing the keyboard. It is about using the right input method for each context. For natural language content — emails, messages, documentation, notes — voice input with AI enrichment is measurably faster and produces better first drafts.

The key enabler is seamless integration. If voice input requires opening a separate app, recording, transcribing, copying, and pasting, the overhead kills the speed advantage. Push-to-talk with direct cursor insertion eliminates that friction.