How can we help?
Answers to common questions below.
How do I clone my voice?
On first launch, after entering your name, Parla asks for a 30-second voice sample. Read the on-screen passage at your normal pace — accent and all. Your voice clone is ready in seconds. You can re-record it any time from the Profile screen.
Where is my voice clone stored?
Your voice model is stored by ElevenLabs, our voice synthesis provider. We do not store the clone on Parla servers. You can delete the clone permanently at any time from the Profile screen.
How does the AI coach generate feedback?
After every recording, your audio is transcribed by OpenAI, then your transcript is sent to a coaching model that scores clarity, pacing, and filler usage. The model also rewrites your answer the way a confident speaker would say it. Improved audio is generated using your cloned voice (or a default coach voice).
Can I cancel my subscription?
Yes — manage your subscription from the iOS Settings → Apple ID → Subscriptions menu. All Parla subscriptions are billed by Apple; refunds and cancellations follow Apple's standard policy.
Is my data private?
Recordings are processed once and not stored long-term on Parla servers. Progress data (XP, streaks, lessons) lives only on your device. See the Privacy Policy for the full breakdown.
Does Parla support languages other than English?
Yes — voice cloning and voiceover output work across 32 languages. Speech coaching feedback is currently optimized for English; we're adding other languages.
I'm having trouble with the microphone.
Make sure Parla has microphone permission in iOS Settings → Privacy & Security → Microphone. If recordings come through quiet, the in-app meter shows when your level is good — try moving closer to the mic or speaking up.
How do I report a bug or request a feature?
Reach out through the in-app feedback flow on the Profile screen. Include a short description and a screenshot if you can.