Drop in a recording and get a clean, accurate transcript in seconds. No sign-up, no software, no watermark.
This free audio to text converter turns any recording into readable, editable text. Drop in
a lecture, interview, voice memo, meeting recording, or the audio from a video, and you get
a full transcript back in seconds. You can switch to a timestamped view to jump to any
moment, copy the whole thing to your clipboard, or download it as a plain .txt file.
It is built for the people who are tired of retyping recordings by hand: students turning lectures into study notes, journalists working through interviews, podcasters drafting show notes, and anyone who needs a searchable record of what was said. Because language is detected automatically, you do not have to configure anything. Upload and read.
Researchers pull quotes from field recordings, support teams keep a searchable log of their calls, and anyone with a phone full of voice memos can finally read them back instead of scrubbing through audio. Accuracy tracks the quality of the recording, so a close, quiet capture reads almost word for word, while a distant or noisy one usually needs a light edit afterward. Whatever you upload stays yours: every file sits behind a private access token that only you receive, and it is deleted automatically after 30 days.
A few honest limits. This browser tool works best on clear speech, so if your recording is noisy, run it through our background noise remover or audio normalizer first for a cleaner result. Need subtitles instead of plain text? The SRT generator produces timestamped caption files. Recording something new? Check your gear first with the mic test.
Files are capped at 100 MB and free use is limited to 30 requests per hour, which is plenty for everyday transcription. Need longer files, batch processing, or higher limits? See pricing.
Accuracy depends on the recording. Clear speech in a quiet room lands very close to word-perfect, because we route your file through leading speech engines like Whisper and ElevenLabs Scribe. Heavy accents, crosstalk, background noise, or muffled audio will lower accuracy, so it is worth cleaning the file first with our noise removal tool.
You can drop MP3, WAV, M4A, AAC, FLAC, OGG, and most common video files like MP4 and MOV. We extract the audio track automatically, so you do not need to convert anything yourself. If a format ever fails, transcode it to MP3 or WAV and try again.
The free browser tool accepts files up to 100 MB, which is roughly two hours of typical spoken MP3. Longer or larger files need to be split first, or handled through our API for batch and long-form work. See the pricing page for higher limits.
Yes. Your file is uploaded over an encrypted connection, transcribed, and then held behind a private access token that only you receive. Files are automatically deleted after 30 days, and we never sell your recordings or use them to train models.
The browser tool is free because it is a taste of the SoundHalo audio API that developers pay for. Free use is capped at 30 requests per hour, which comfortably covers personal transcription without an account. Teams and automated workloads move to the paid API.
Yes. Language is detected automatically, so you can upload Spanish, English, Portuguese, French, German, and dozens of other languages without changing any settings. If you mainly work in Spanish, try our Spanish version at Convertir Audio a Texto.