Support & FAQ

Every paid request has an explicit operation. Fix & Enhance sends the recording through GPU speech repair, then uses measured audio evidence to plan and render a mastering chain. Ask searches the DSP knowledge base and cites matching passages when available. Paid work runs as a durable backend job, so leaving a screen or briefly backgrounding the App does not cancel the server-side run.

App Store builds use AuraLinter's managed service, currently at https://auralinter.fly.dev. The service handles model inference, GPU processing, identity, billing, and the PostgreSQL database. Custom backend controls exist only in authorized developer builds and are not a supported consumer feature. Publication of a source repository does not itself grant a license to operate or modify that code.

The managed consumer service uses Claude Sonnet 4.6 through Amazon Bedrock for grounded Ask answers and for the mastering planner behind Fix & Enhance. Speech enhancement runs separately on dedicated audio models. Public App requests are explicitly Ask or Fix & Enhance; they are not routed to the backend's internal code-generation/evaluation path. Authorized developer deployments can configure supported provider and model IDs.

AuraLinter's RAG system uses hybrid search over a pgvector-enabled PostgreSQL database. It combines two retrieval strategies and merges them with Reciprocal Rank Fusion (RRF):

Dense (semantic): HNSW-indexed vector search using Voyage AI embeddings (1024-dim). Finds conceptually similar passages even with different keywords.
Sparse (keyword): PostgreSQL full-text search via ts_rank, weighted over title and body columns. Matches exact terminology.

When retrieval finds relevant passages in the ingested DSP corpus (textbooks, papers, technical notes), the answer cites them inline as [1], [2] so you can see which claims are grounded in the corpus rather than in general knowledge. The corpus documents are internal reference material, so the app shows the markers but does not publish the underlying source list. If no passage matches, AuraLinter labels the answer as general knowledge rather than fabricating a citation.

Audio is uploaded to the backend and staged in temporary files while it is analyzed or enhanced. Synchronous-session files normally expire after about 30 minutes; long-job files may remain for up to about four hours; a finished full-episode enhancement is kept for up to about 24 hours so the App can download it. Saved projects and media remain in the App's private storage until you delete the project or remove the App. Completed run records are stored separately. See the Privacy Policy for full details.

AuraLinter requires iOS 17.0 or later and runs on any iPhone or iPad that supports iOS 17. The app uses SwiftUI, AVAudioEngine, URLSession, and a bundled Xiph codec library whose license is shown in the App. Compute-intensive work (LLM inference, vector search, and neural speech enhancement) runs on the backend, so the iOS app itself remains lightweight. An active internet connection is required for paid cloud features.

Not as a supported App Store feature. App Store builds use the managed service and do not expose backend controls. A source repository may be visible for engineering transparency, but source availability is not a license grant. Do not copy, modify, redistribute, or operate it unless an express license file or written permission authorizes that use.

App Store builds automatically create an anonymous per-install identity protected by Apple App Attest. Its credential is stored in the iOS Keychain and is used to scope projects, subscription status, allowances, and service requests to that installation; there is no API-key prompt or user-managed service password. Production builds do not expose backend URL or credential editing. Authorized developer builds can configure development backends and authentication; unauthenticated mode is appropriate only for isolated local development.

Rate limits are configurable via backend environment variables. The defaults include:

Rate limiting: Enabled via RATE_LIMIT_ENABLED=true. Limits requests per client per time window to prevent abuse.
Paid allowances: the App shows the Fix & Enhance and Ask allowances included with the active subscription period.
Audio duration: the sync Fix & Enhance lane accepts clips up to 5 minutes. Episode jobs accept recordings up to 60 minutes.
Episode fair use: episode checks use no per-use credits. Episode enhancement is limited to 3 nonfailed submissions per rolling 24 hours. Server admission controls may also return a retry response while audio capacity is busy.

Managed-service limits are enforced server-side and may change to protect reliability, safety, or provider budgets; material subscription limits are disclosed on the purchase surface.

Yes. Neural enhancement and automated mastering can introduce artifacts or make a creative choice you do not prefer. AuraLinter keeps the original, presents an Original/Improved A/B, and reports measured changes so you can judge the result before export. Listen at a safe level and keep the original recording for any important project.

The iOS app does not include third-party advertising or behavioral analytics SDKs. It sends audio, prompts, and service activity to the managed backend to provide the requested features. If a video is attached to Ask, it also sends up to three sampled JPEG stills with the question; the original video file is not uploaded. Saved project media remains local, and credentials are stored in the device Keychain. See our Privacy Policy for storage and retention details.

Use the Contact page to reach us. Select "Bug report" for app issues (please include iOS version, device model, and steps to reproduce) or "General inquiry" for questions. The contact form includes a verification slider to prevent spam — slide it fully to unlock the submit button.