錄製語音，說話時即時顯示轉錄。
Click Record and grant mic access.
Transcript appears as you speak.
Download audio (.webm) and transcript (.txt).
Save as WebM audio file.
Speech-to-text appears as you speak.
English, Spanish, French, German, Japanese, etc.
Audio and transcript download separately.
A dictaphone records voice for later transcription. Modern browsers can do the transcription in real time, eliminating the typing step entirely. Speak your thoughts, walk away, come back to a finished transcript. Uses: meeting notes, journalism interviews, voice memos, lecture notes, brainstorming sessions, blog drafts, podcast scripts. Anywhere typing slows you down, dictation accelerates.
Chrome and Edge include the Web Speech API, which sends audio to Google's speech recognition servers in real time and returns text. The Web Speech API supports 100+ languages and handles accents, dialects, and varying speech rates surprisingly well. Firefox doesn't implement the Web Speech API. On Firefox, our dictaphone still records audio but transcription isn't available: you'd need to use a separate service like Whisper or Otter for transcription afterwards.
We support 13 common languages in the dropdown: English (US/UK), Spanish, French, German, Italian, Portuguese, Russian, Chinese, Japanese, Korean, Arabic, Hindi. Accuracy varies by language: English is highest (>95% for clear speech), followed by major European languages. Less-supported languages may have lower accuracy. For best results: speak clearly and pause between sentences. Background noise hurts accuracy significantly. Cheap headset mics often outperform built-in laptop mics for transcription clarity.
Meeting recording: capture both audio and an automatic transcript of meetings. Search the transcript for key points later. Interviews for journalism: record + transcribe interviews simultaneously. Saves hours of manual transcription. Lecture notes: students record lectures and let transcription handle the writing. Search the transcript for specific topics. Voice journaling: speak your thoughts; review the transcript later. Faster than typing for many people. Podcast preparation: dictate your script ideas; the transcript becomes your first draft.
Audio recording: stays in your browser memory only. Nothing uploaded to our servers. Live transcription: Chrome and Edge send audio to Google's speech recognition servers. This is required for the transcription to work. If privacy is critical, use a different tool (or our regular Voice Recorder) and transcribe later with a local model. Don't use this for highly sensitive content unless you've confirmed your transcription pipeline meets your privacy needs.
適用於所有主流平台和瀏覽器
| 裝置 / 系統 | Chrome | Firefox | Safari | Edge |
|---|---|---|---|---|
| Windows 10/11 | ✅ | ✅ | : | ✅ |
| macOS Ventura+ | ✅ | ✅ | ✅ | ✅ |
| Android 8+ | ✅ | ✅ | : | ✅ |
| iPhone / iPad (iOS 14+) | ✅ | : | ✅ | : |
| Chromebook | ✅ | : | : | : |
| Linux (Ubuntu) | ✅ | ✅ | : | ✅ |
為此工具評分
4.6 分(滿分 5)· 2,517 則評分