NOTE: This page is for Soundtrap for Music Makers (personal accounts). It does not apply to Soundtrap for Education. Schools: see Machine learning in Soundtrap for Education, which describes the features, access and data protection that apply to school accounts.
Soundtrap uses machine learning (sometimes referred to as AI) in a small number of optional tools for creating music and finding sounds.
- No chatbots or conversational AI (except support tickets)
- No automated grading, student profiling, or decision-making about students
- No use of student or school data to train AI models
Machine learning features
| Feature | What it does | Available on | Model, where it runs |
|---|---|---|---|
| Transcription | Speech to editable text | All plans | Google Speech-to-Text (Google's service) |
| Noise reduction | Removes background noise | All plans | RNNoise, run by Soundtrap |
| Generate Drums | Creates a drum pattern | All plans | Google Magenta, in the browser |
| Key detection | Finds a recording's key | Music Production, Vocals & Songwriting, and Production & Vocals | Spotify Basic Pitch, in the browser |
| Stem Splitter | Splits a song into vocals, drums, bass, other | Production & Vocals | Meta Demucs, run by Soundtrap |
| Top Picks | Suggests loops that fit the loops already in the project | All plans | Soundtrap's own model, run by Soundtrap |
| Loop search | Finds loops that match what the user types | All plans | Google Gemini (Google's service) |
All Soundtrap services are hosted on Google Cloud. "Google's service" means the data is processed by Google's own AI model; "run by Soundtrap" means only Soundtrap's code processes it.
Output is saved in the user's project like any other content. Users can edit or delete it at
any time.
How your data is handled
-
Only when you use a tool. Machine learning runs only when you use a tool, such as
opening the loop browser or searching for loops. Nothing processes your audio
automatically. -
Limited output. No tool generates open-ended content. Each tool returns a specific
result: text from your own speech, cleaned or separated audio, drum notes, a musical
key, or a list of loops from Soundtrap's sound library. -
Your control. All output is saved in your project and can be edited or deleted.
Transcripts are fully editable to correct mistakes. -
Training. Your recordings and projects are not used to train AI models. Top Picks
learns which loops work well together from how Soundtrap users combine loops in
their projects; it uses only which loops are combined, not your audio or account
details. -
Search quality. A small sample of loop search phrases, de-identified and without
account details, is used to measure and improve search results. -
Google Cloud. Transcription audio and loop search text are processed by Google
Cloud, and no account details are sent. Google does not use this data to train its
models. All other tools run on Soundtrap's servers or in your browser. - Temporary files. Temporary processing files are deleted automatically within 7 days.
For more about how Soundtrap handles personal data, see our Privacy Policy.
Comments
0 comments
Article is closed for comments.