How to Transcribe Audio to Text Free (Voice Notes, Lectures, Video)

By HSP · Published October 21, 2026 · 3 min read · Writing & Text

Typing up a recording by hand is slow: a ten-minute voice note can take most of an hour. Speech recognition has become good enough that a computer can do the first draft for you, and you can now do it for free without sending your audio to anyone.

This guide uses our free Audio to Text converter, which runs Whisper, the open-source speech recognition model released by OpenAI. According to OpenAI, Whisper was trained on 680,000 hours of audio and transcripts collected from the internet, and it handles many languages as well as translation into English. The model runs on your own phone or computer.

Step 1: Get your audio ready

• Voice notes and recordings: share or save them as a file (M4A on iPhone, often OGG or M4A on Android).
• Videos: you can use the MP4, MOV or WebM file directly; the tool reads its sound track.
• Live speech: press Record from microphone on the page.

Step 2: Choose the model and language

Fast (Whisper Tiny, about 41 MB) is quick and fine for clear speech. Accurate (Whisper Base, about 77 MB) makes fewer mistakes with accents and noise. The model downloads once and is then saved by your browser. Leave the language on Detect automatically, or set it yourself, for example Hindi or Punjabi, for better results. Choose Translate to English if you want an English version of speech in another language.

Step 3: Transcribe, check and export

Press Transcribe and keep the tab open. Laptops are usually faster than phones. When it finishes, read through the text in the box and fix names, numbers and technical words, which every speech system can mishear. Then copy it, or download a TXT file for notes, or SRT/VTT subtitle files with timings for YouTube and video editors.

Tips for a more accurate transcript

• Record close to the speaker, in a quiet room, without music in the background.
• Ask people not to talk over each other.
• For long lectures, the Accurate model is worth the extra download.
• Choose the language yourself when the speaker mixes languages.
• After transcribing, use our text summarizer to pull out key points, and the word counter to check length.

Uses people love

• Students turning recorded lectures into revision notes.
• Creators making subtitles, which help viewers who watch with the sound off.
• Journalists and researchers transcribing interviews.
• Families saving a grandparent's voice message as text.

Need live dictation instead of a file? Our speech to text page types as you talk.

Privacy and consent

Because the tool runs on your device, private recordings never leave it. Still, always ask before recording other people, and follow your school's or workplace's rules on recording lessons and meetings.

VSH take

Most free transcription sites upload your file and stop after a few minutes. Running Whisper in the browser flips that: no limit, nothing uploaded, and good enough accuracy for a first draft. Treat the result as a draft and give it a quick proofread, especially for names and numbers.