Free tool

Free, no sign-up: upload an MP3, M4A, WAV or other audio file up to 10 minutes and 500 MB, and ClipSDK transcribes it with timestamps, ready to copy or save as TXT, SRT, VTT or JSON, 3 files a day.

Audio to text converter: transcribe audio files for free

Podcasts, voice memos, interviews, lectures: ClipSDK takes recordings up to 10 minutes, or 60 with a free account, writes them out with Whisper in lines of at most 6 seconds, and plays your file from any line you click.

Transcribe an audio file

Runs on ClipSDK servers

No sign-up.

Your file goes to ClipSDK servers, which transcribe it and delete the upload and the transcript 24 hours later.

Free toolsUpdated

Free audio transcription next to Otter and Riverside

What each free option gives you, as the vendor's own page puts it.

Free audio transcription next to Otter and Riverside
Free allowanceLongest recordingDownloads
ClipSDK3 files a day, no sign-up; 10 with a free account10 minutes, or 60 with a free account; 500 MBTXT, SRT, VTT and JSON
Otter Basic300 minutes a month, 3 file imports in total30 minutes per conversationNot listed on its pricing page
RiversideFree, no limit statedNot statedSRT and TXT

Transcripts through the ClipSDK API

POST /v1/clips takes a video by link and transcribes it with Whisper as the first step of every job. It returns clips rather than a full transcript: each clip carries its own text plus SRT and VTT files.

This request ran on a 3-minute sermon recording: 2 clips with karaoke captions came back in 63 seconds, each with its transcript, SRT and VTT. At $0.03 per source minute that's $0.09 (pricing).

POST/v1/clipsTry it
curl https://api.clipsdk.com/v1/clips \
  -H "Authorization: Bearer $CLIPSDK_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "video_url": "https://api.clipsdk.com/f/lib/demo/sermon-at-calvary.mp4",
    "clips": 2,
    "max_seconds": 45,
    "aspect": "9:16",
    "reframe": "auto",
    "captions": "karaoke"
  }'

How to transcribe an audio file

Upload your recording in the tool above. ClipSDK uses Whisper to turn the speech into text with timestamps. You don't need to sign up. You can use it for a podcast, an interview, a lecture, a voice memo or a recorded call.

Your file plays on the page, right above the transcript. Click a timestamp to hear the recording from that line. The text is split into lines of at most 2 rows of 42 characters, with each line covering no more than 6 seconds. These are the same lines you'll get in the SRT and VTT files.

No speech in the file? If the recording has no sound or no speech, the run stops and tells you. If you refresh the page, the link in your address bar picks up the same run again.

Formats, length and languages

You can upload MP3, M4A, WAV, OGG, Opus, FLAC and AAC files. Video files work too. The file needs to be within both limits: 10 minutes long and 500 MB in size. With a free account, it can run up to 60 minutes.

A recording over 10 minutes waits for you to sign up free, then its transcript starts by itself. Recordings over 60 minutes won't be accepted, so trim a longer recording before uploading it. For video files and YouTube or TikTok links, you can also use the video to text converter. It does the same job.

You don't need to choose a language. ClipSDK detects it from the speech, and the transcript stays in that language. It doesn't translate what's said into another language.

What to do with the transcript

You can copy the whole transcript as text, or copy it with timestamps included. If you want a file, download TXT for plain text, or SRT or VTT for subtitles.

There's also a JSON download for developers. It contains every line plus word-by-word timings. The transcript doesn't label speakers, though: it's one stream of timed lines, including when different people speak.

Only need the gist? The free AI video summarizer turns a recording into a short summary with key points. And if you're choosing what to share from an episode, the clip finder picks the best moments from a transcript.

Questions, answered

Is this audio to text converter free?

Yes, it's free, with no sign-up. Everyone gets 3 free runs a day per network, meaning a home or office connection, and those runs come back at midnight UTC. A free account gets 10 runs a day of its own.

Which audio formats does it take?

It takes MP3, M4A, WAV, OGG, Opus, FLAC and AAC files. You can upload video files too.

How long can the recording be?

Up to 10 minutes without an account, or 60 minutes with a free one, and the file must also be no larger than 500 MB. Anything over 60 minutes is refused, so you'll need to trim it first.

Can it transcribe audio in other languages?

Yes, it detects the spoken language on its own. The transcript comes out in that language, not as a translation.

Does it tell speakers apart?

No, it doesn't add speaker labels. You'll get one stream of timed lines rather than text divided up by who's talking.

Is my audio kept?

The run and its files are deleted 24 hours after the run starts. Until then, the run's link stays in your address bar, and refreshing the page picks it up again.

Start free

Your back catalog is full of shorts

Your first $1 is free: 33 minutes of video, every clip included up to 1080p60 and 20 Mb/s.

Two studio microphones on a desk in a dark room lit by a red glow