SoundCloud hosts audio and nothing else, so there is no caption file to lift: paste a track link and JotVox transcribes the sound itself.
Get a SoundCloud transcript →Short answer: no. SoundCloud is a place to publish sound. A track page carries a waveform, a description, tags and a comment stream, but nothing that stores the words being spoken. There is no caption file sitting behind the player waiting to be pulled.
That flips the usual order of operations. On a video site like YouTube, the first move is to check for existing captions and fall back to AI only when there are none. On SoundCloud, AI transcription is the first move and the only one. JotVox sends the audio through its speech model, which covers 99 languages and returns text at 98%+ accuracy on clear recordings.
It also means every SoundCloud job uses a credit. New accounts get three free credits plus one free AI transcription a day once your email is verified, which is enough to run a few interviews or a lecture before you decide whether to buy a pack.
This is the question that decides whether a transcript is worth a credit, and SoundCloud is the one platform where it genuinely matters. A huge share of what gets uploaded is music. A song has lyrics, not speech, and lyrics buried under drums, bass and reverb are hard for any transcription system, human or machine.
Where SoundCloud transcription works well:
Where it works less well: dense instrumental music, tracks with heavy vocal processing, and long DJ mixes where the only speech is a shout-out every twenty minutes. You will still get something back from a DJ set, but most of it will be gaps and fragments of whatever was audible over the beat. If the same set also went out live, a saved Twitch broadcast of it is sometimes the better source.
SoundCloud is full of audio that was never recorded in a studio. Phone recordings of panel discussions. Interviews over a bad connection. Live sets captured on a single room mic. The model copes with more of this than people expect, but accuracy tracks the recording: two people speaking clearly into their own mics come back close to perfect, while a crowded room with an echo does not.
Length is the other SoundCloud quirk. Uploads run from a ninety second voice memo to a three hour mix. One credit is one action, so a long file is one transcription, not one charge per hour. What changes with length is what you do next. Past about an hour you want an AI summary or chapter breakdown so you can find the part you actually need, and that is a separate action on its own credit.
Journalists pull quotes from interviews that were published as audio and never written up anywhere. Students turn a lecturer's uploaded recording into notes they can search during revision. Comedians and spoken word performers get a text version of a set so they can cut and rework the material. Researchers working with oral history collections, which land on SoundCloud more often than you would guess, get something they can quote and cite.
Once the text exists, exports cover the usual handoffs: TXT for a document, SRT or VTT if the audio is going into a video edit, and MP3 if you want the sound file alongside the words. If you want the same treatment on episodes published through a podcast app rather than SoundCloud, the Apple Podcasts page covers that route.
Private and unlisted tracks. A private SoundCloud track is not publicly readable, so a link on its own will not give JotVox access to the audio. If the track is yours, download the file and upload it directly, or from Google Drive or Dropbox.
Region-blocked tracks. Some uploads are unavailable in certain territories, usually for rights reasons. Audio that cannot be reached cannot be transcribed.
Deleted or taken-down tracks. Uploads disappear from SoundCloud over rights claims fairly often, so a link sitting in your bookmarks may point at nothing.
Live audio still in progress. Transcription runs on a finished file, so wait until the recording is published. In all of these cases, uploading the audio file yourself is the reliable path. See also transcribing MP3 and MP4 files directly.
1. Copy the track link Open the track on SoundCloud in a browser or the app. Use the share option on the track itself and copy the link, or copy the URL straight from your browser's address bar.
2. Paste it into JotVox Drop the link into the box at jotvox.app. If the track is private or unavailable, upload the audio file instead from your device, Google Drive or Dropbox.
3. Run the AI transcription SoundCloud has no captions to grab, so JotVox transcribes the audio. Let it detect the language automatically across 99 languages, or set it yourself. This uses one credit.
4. Export or summarise Download the result as TXT, SRT, VTT or MP3. For a long mix or a full interview, run an AI summary first to get key takeaways and chapters before you read the whole thing.
Up to a point. SoundCloud tracks have no caption file, so every transcript needs AI transcription, which uses credits. A verified new account starts with three free credits plus one free AI transcription each day, so you can do roughly a track a day at no cost. After that, credit packs start at $4.99 for 20 credits and subscriptions start at $9 a month.
It will try, but music is the hardest case on SoundCloud. Speech models are trained on speech, and sung vocals sitting inside a mix of instruments, effects and reverb often come back partial or wrong. A credit is well spent on interviews, lectures, talks and spoken word sets. On a dense instrumental track or a club mix, expect very little usable text.
Long uploads are fine, including multi-hour DJ sets and full conference recordings. One credit is one action, so a three hour file is a single transcription rather than a charge per hour. The practical limit is what you do with the result: past about an hour, run an AI summary or chapter breakdown so you can jump to the section you need instead of scrolling through thousands of words.
Not from the link alone. Private tracks are not publicly accessible, so JotVox cannot fetch the audio the way it can from a public URL. If the track is yours, download the original file from SoundCloud and upload it to JotVox from your device, Google Drive or Dropbox. That path runs the same transcription and produces the same transcript.
It is processed in memory and deleted right after the transcription finishes. JotVox keeps no copy of the SoundCloud audio it fetched or the file you uploaded. You keep the transcript and any exports you download. That matters if you are working with unreleased material, interview recordings under embargo, or anything else you would rather not leave sitting on a server.