Audio to Text Converter

The Audio to Text Converter turns speech into text: a voice memo, an interview, a lecture, a meeting recording, a podcast episode, a phone call or the sound track of a video. Choose a file or record with your microphone, and the transcript appears line by line, each line with the time it was said.

Two ways to transcribe

What you can do with the transcript

• Check it against the recording: click any timestamp to hear that moment, and click any line to correct it.
• Copy or download it as plain text, with or without timestamps and speaker names.
• Make subtitles: download SRT or VTT captions, timed to the speech and split into readable lines.
• Keep working with it: translate it, summarize it into key points, or burn the subtitles into your video.

Tips for a good transcript

• Record close to the speaker, away from music and fans; noise costs more accuracy than anything else.
• For interviews and meetings use AI mode, so each line says who spoke.
• If a video will not open, extract its sound with the Audio Extractor and transcribe the MP3.

Frequently asked questions

How do I convert audio to text?

Choose the audio file (or record with the microphone), then press Transcribe free or Transcribe with AI. The text appears line by line with a timestamp on each line. Copy it, or download it as a TXT file or as SRT or VTT subtitles.

Is it really free?

Yes. The free mode runs a speech recognition model (OpenAI's Whisper) inside your own browser: every visitor gets 10 free minutes of it a day (each file counts at least 1 minute), Pro members get 200 minutes a day, and with Premium it is unlimited. AI mode costs 3 credits per minute of speech, and every visitor gets 3 free AI credits a day, so the first minute of AI transcription each day costs nothing.

What is the difference between the free mode and AI mode?

The free mode is private (10 minutes a day free, 200 with Pro, unlimited with Premium) and does well on clear speech in one language, but it is less accurate on accents, noise and languages other than English, and it does not tell speakers apart. AI mode adds proper punctuation, detects the language by itself, copes better with accents and background noise, and labels each speaker (Speaker 1, Speaker 2), which you can rename.

Can I convert a video to text?

Yes. Choose the video file (MP4, MOV, WebM, MKV and others) and the tool reads its sound track. In AI mode only the sound is sent, as a small compressed audio file; the video itself never leaves your computer.

Can I make subtitles from it?

Yes. Download the transcript as SRT or VTT: each line becomes a subtitle with its start and end time, long lines are split into readable two-line captions, and you can open it in YouTube, Premiere, CapCut or VLC, or burn it into the video with our Add Subtitles to Video tool.

Which languages are supported?

AI mode transcribes more than 100 languages and detects the language by itself, including English, Spanish, Portuguese, French, German, Chinese, Japanese, Korean, Russian, Arabic and Hindi. The free mode supports the same main languages, but you choose the spoken language first.

How long can the recording be?

Up to 3 hours or 500 MB. In the free mode, visitors without a plan can transcribe 10 minutes a day, Pro members 200 minutes, and Premium members have no daily limit. Long recordings are processed in parts of up to two minutes, cut at pauses so no word is split, and the transcript grows on the screen as each part is done. You can stop at any time and continue later.

Is my recording private?

In the free mode nothing leaves your computer. In AI mode the sound is sent in parts to be transcribed and is not stored; the transcript is shown only to you and is not kept.

How accurate is it?

On clear speech, AI mode typically gets well over 90% of words right, with line times accurate to a fraction of a second. Names, technical terms and quiet or overlapping speech are where mistakes happen, so every line can be corrected on the page before you download it.

Can I transcribe an iPhone voice memo or a WhatsApp voice message?

Yes. Voice Memos saves .m4a files and WhatsApp voice messages are .opus or .ogg files, and both are accepted here. Save or share the recording to your files first, then choose it with the upload button. A short memo fits easily in the 10 free minutes a day.

Can it transcribe a Zoom, Teams or Google Meet recording?

Yes. Choose the recording file: Zoom saves an MP4 video and, if you allow it, a separate M4A audio file, which is smaller and quicker to read; Teams and Meet save MP4. AI mode labels each speaker, and you can type in their real names before you download, so the transcript reads like meeting notes.

How many credits does a one-hour recording cost?

AI mode costs 3 credits per minute of speech, so one hour is about 180 credits; parts that are mostly silence are not charged. The free mode uses no credits at all: 10 minutes a day without a plan, 200 with Pro, and no limit with Premium.

Can I get the transcript of a YouTube video?

If the video has captions, use our free YouTube Transcript Generator: paste the link and it shows the video's own captions as text, with no download and no credits. If you have the video or audio file yourself, choose it here to transcribe it.

Can I download the transcript as a Word document?

Yes. Press DOC to download a file that opens in Microsoft Word, Google Docs or LibreOffice. Each speaker's turn is one paragraph with the speaker's name in bold, and the times are included when Timestamps is ticked. TXT, SRT and VTT downloads are there as well.