Audio to Text Converter, free and online
Upload MP3, M4A, WAV, FLAC or any other common audio file. Get a transcript with timestamps, then summarize, translate or export it.
- Free, no signup for files up to 25 MB
- 36 languages with auto-detect and translation
- Whisper large-v3, AI summary and chat
Drop audio or video here
MP3, MP4, WAV, M4A, OGG, WEBM, FLAC
What you get
An example result for a short interview. Click through the tabs.
- 0:00
- Hi, thanks for making time today. Let us start simple.
- 0:07
- I started making short films back in college as a hobby.
- 0:15
- What draws you to a story before you commit to it?
- 0:24
- Real people, real moments. I observe first, script later.
- 0:34
- How do you keep the crew energy up on long shoots?
- 0:42
- Small rituals. Coffee at call time, a walk before magic hour.
- 0:54
- What advice would you give someone starting out?
- 1:02
- Watch a lot. Then shoot even more. Nothing beats reps.
Illustrative sample. Real transcripts have no speaker labels.
Supported audio formats
No converting first. Pick your format to see what we accept and how much fits in the free limits.
MP3
The default for recorders, podcast downloads and files shared over chat.
- Usually from
- Recorders, podcasts, chat downloads
- We accept
- Any bitrate, mono or stereo, CBR or VBR
- BitrateFor speech, more than about 64 kbps stops helping. The speech model works at 16 kHz mono internally.
- Variable bitrateVBR files sometimes show the wrong length in older players. The transcript and its timestamps are not affected.
- ID3 tagsIgnored. You get the spoken words back as text.
M4A
AAC audio in an MP4 container. This is what an iPhone records.
- Usually from
- iPhone Voice Memos, GarageBand, FaceTime
- We accept
- AAC in an MP4 container
- From an iPhoneIn Voice Memos tap Share, save the memo to Files, then upload it here. No conversion needed.
- Long memosVoice memos are small, so the length limit comes first: 30 minutes as a guest, 180 with a free account.
WAV
Uncompressed audio. The cleanest source and the biggest files.
- Usually from
- Field recorders, DAW exports, Windows Voice Recorder
- We accept
- PCM WAV, mono or stereo, any sample rate
- About 10 MB a minuteAt CD quality WAV fills the guest limit in about two and a half minutes.
- Shrink it firstExport it as a mono MP3 at 64 kbps: nothing is lost for speech and the file gets more than ten times smaller.
- Big filesWith a free account, files over 60 MB are split into up to 3 parts and joined back into one transcript.
FLAC
Lossless and roughly half the size of the same WAV.
- Usually from
- Music and oral-history archives, field recording
- We accept
- FLAC, mono or stereo
- Worth the size?For accuracy, no. FLAC and a decent MP3 transcribe the same. Keep FLAC as the archive copy and upload an MP3 if the file is too big.
OGG
An open container, usually with Opus audio. Voice messages use it.
- Usually from
- WhatsApp and Telegram voice messages, open-source recorders
- We accept
- OGG with Opus or Vorbis, plus .opus and .oga files
- Voice messagesTelegram .oga and WhatsApp .opus files upload as they are, no renaming needed.
AAC
The same codec as M4A, sometimes saved with an .aac extension.
- Usually from
- Android recorders, some screen recorders
- We accept
- AAC in an MP4 container
- Raw AAC streamsSome Android recorders save raw ADTS AAC, which the speech model cannot read. If the upload is rejected, convert the file to MP3 first.
WebM
What browsers and screen recorders produce, usually with Opus audio.
- Usually from
- Browser recordings, OBS, Loom, Google Meet
- We accept
- WebM with Opus or Vorbis audio
- Video inside?We take the audio track and ignore the video. A WebM screen recording is mostly video, so the file can be big for little audio.
Also accepted: MP4, MOV and MPEG video, where we use the audio track. AIFF, WMA and AMR need converting to MP3 first. Sizes are approximate and depend on the bitrate.
Audio quality matters more than format
Clear speech
One voice close to the microphone, little background noise. This gives the most accurate transcript.
Noisy audio
Music, traffic or people talking over each other lower accuracy in any format.
Heavy compression
Very low bitrates, under about 32 kbps, can blur the sounds the speech model needs.
Tip: for speech, 64 kbps mono is enough. A higher bitrate makes the file bigger, not the transcript better.
Transcribe audio for
Frequently asked questions
What audio formats can I upload?+
MP3, M4A, WAV, FLAC, OGG (including .opus and .oga voice messages), AAC and WebM, plus MP4, MOV and MPEG video, where we use the audio track. AIFF, WMA and AMR files need converting to MP3 first.
How long does transcription take?+
A 10-minute file usually takes 15 to 30 seconds and a 60-minute recording about a minute. Mictoo is free and runs on a limited daily transcription budget, so at busy times your file waits in a queue and the page shows your position and an estimated wait.
Is audio transcription really free?+
Yes. Without an account you can transcribe 3 files every 24 hours, up to 25 MB and 30 minutes each, with AI summary, translation and TXT or SRT export. A free account allows 7 files every 24 hours (up to 5 an hour), up to 180 MB and 180 minutes each, and adds chat, a 7-day history and more export formats. No credit card.
Are my audio files stored?+
Your file goes to temporary storage and is deleted once the transcript is ready. Anything left behind is removed within about two hours. Guest transcripts are not saved on our servers. If you are signed in, transcripts stay in your History for 7 days and are then deleted.
Does the file format affect accuracy?+
Hardly. Audio quality matters far more: clear speech in a small MP3 beats a noisy WAV. Every format goes through the same Whisper large-v3 model.
Do I need to convert my audio first?+
Not for the formats above: upload the file as it is. Only AIFF, WMA, AMR and raw AAC streams from some Android recorders need converting to MP3.
Can I edit the transcript before downloading?+
Yes. Switch to the Editor view to fix names and specialist terms, and use search to find every place a word appears. Then copy or download the text.
How many languages are supported?+
Leave the language on auto-detect, or pick one of 36 languages from the list if the recording is short or mixes languages.
Can I translate the transcript?+
Yes, into any of the same 36 languages. Translation works without an account.
What export formats are available?+
Everyone can download TXT, TXT with timestamps and SRT subtitles. With a free account you can also export DOCX, PDF, VTT and JSON.
Drop any audio file and get the text
Free, no signup, 36 languages with auto-detect.