Last updated: October 12, 2026
We collect information you provide directly to us when you:
We do not sell your personal data. We share data only with:
Underlying model disclosure. Speech recognition on Dictate is performed by the
Volcano Engine Doubao streaming speech recognition model (bigmodel), accessed through the
Volcano Engine Open Speech Platform. Dictate is an independent relay service; we do not present
ourselves as an official product of the model provider.
Audio frames you send through the WebSocket relay are forwarded to Volcano Engine frame-by-frame for recognition and are never stored on Dictate servers: frames are not decoded, not written to disk, and their content is not logged. We record only usage metadata for billing and capacity management — session identifier, duration in milliseconds, frame count, sample rate, the recognition key used, connection status and timestamps. We do not use your audio or transcriptions to train models.
As a result, audio you submit may be processed in servers located outside your own jurisdiction. Our sub-processor list is maintained in our Data Processing Addendum.
We retain your data for as long as your account is active. When you delete your account, we remove or anonymize your personal data within 30 days.
Depending on your jurisdiction, you may have the right to:
You may not use Dictate to transcribe or distribute content in any of the six prohibited categories set out in our Acceptable Use Policy, which also governs our child safety (CSAM) zero-tolerance commitment, reporting channels and enforcement process.
For privacy-related inquiries, contact us at support@dictate.top.