Audio and transcripts
Audio is processed in memory to produce the transcript and is then discarded. We do not store your audio or transcripts, do not log them, and do not use them to train models.
What we keep
For each request we keep its time, the key it used, the length of the audio, the language, the latency and whether it succeeded. This is what the console's usage pages and your balance are built from.
Keys
Keys are stored as one-way hashes, so we cannot read a key back. If you lose one, revoke it and create a new one in the console.
Questions
Questions about data handling: support@p99lab.com.