ECHOSENSE-AI PRIVACY POLICY

EchoSense-AI is an accessibility-focused application that provides experimental visual, document, translation, currency, navigation, safety, and assistant features. This Privacy Policy explains how the EchoSense-AI applications for iOS and Android handle information.

1. INFORMATION ECHOSENSE PROCESSES

Depending on the features you choose to use, EchoSense-AI may process:

• Camera previews and individual camera frames;
• Photos, images, PDFs, and text documents that you select;
• Microphone input and speech-recognition results when you use voice commands or voice chat;
• Text that you type into the app;
• Text recognized from images or documents;
• Translation requests and results;
• Object detections, device depth estimates, and other safety sensor results;
• Prompts, conversation history, and responses generated by language models;
• Model download and application diagnostic information; and
• Settings and preferences that you choose in the app.

EchoSense-AI does not require an EchoSense-AI account to use its on-device features.

2. ON-DEVICE PROCESSING

EchoSense-AI is designed primarily for on-device processing. When an on-device model or local system service is used, camera images, selected content, prompts, recognized text, and model results are processed on the device.

Camera preview frames and analysis snapshots are not intentionally retained by EchoSense-AI after they are no longer needed for the requested feature. Files that you select remain under the control of the operating system and are accessed to perform the action you requested. Temporary files and normal operating-system caches may exist for a limited period and may be managed or removed by the operating system.

Conversation content may remain in application memory while a chat is active. Starting a new chat, closing the relevant session, or terminating the app may clear that in-memory context. EchoSense-AI does not intentionally upload local chat history to an EchoSense-AI-operated server.

3. OPTIONAL ONLINE AI SERVICES

Online analysis is optional. If you connect an online AI provider and request or configure online processing, EchoSense-AI sends the content needed for that request directly to the provider you selected. Depending on the request, this may include an image, document text, typed or spoken prompts, recent conversation context, and relevant sensor observations.

The currently supported providers are Google LLC (Gemini) and OpenAI, L.L.C. Content sent to a provider is governed by that provider's terms, privacy policy, account settings, and data-retention practices. The provider may charge your account for use of its service.

On iOS, before EchoSense-AI contacts an online AI provider, the app identifies the selected provider, lists the categories of data that may be sent, explains the purpose, and asks for explicit permission. Declining does not prevent use of on-device analysis. You can withdraw permission at any time in Settings under Online Analysis. After permission is withdrawn, EchoSense-AI does not send new AI requests to the provider unless you grant permission again.

If you choose to obtain an API key, EchoSense-AI opens the selected provider's website only after this disclosure is accepted. The provider and its website or content-delivery services receive standard network information such as your IP address, device or browser information, request time, and pages requested. Account details or other information you enter there are provided directly to the selected provider and are governed by that provider's privacy policy.

EchoSense-AI stores a user-provided provider credential using the platform's secure credential storage, such as iOS Keychain or Android encrypted credential storage. The credential is sent to the selected provider only for authentication. It is not intended to be included in application logs or committed to the EchoSense-AI source repository. You can remove the online connection and its stored credential from the app's Settings.

4. CAMERA, MICROPHONE, PHOTOS, AND FILES

EchoSense-AI requests access to the camera to provide visual analysis, document, currency, navigation, safety, and assistant features. Microphone and speech recognition access is requested when you choose voice input. Photo-library or file access is used only when you choose content to open or attach.

Operating-system speech recognition may process audio on the device or through services operated by Apple, Google, the device manufacturer, or the configured speech-service provider. Its behavior depends on the device, operating-system settings, installed services, and network availability.

You can deny or revoke these permissions through the device's system settings, although the related EchoSense-AI features may then be unavailable.

5. SAFETY, OBJECT DETECTION, AND DEPTH SENSING

EchoSense-AI may use the camera, object-detection models, LiDAR or other supported depth sensors, and camera-relative estimates to provide experimental safety information.

On compatible Android devices, depth sensing may use Google Play Services for AR (ARCore). ARCore is provided by Google LLC and is subject to Google's applicable terms and privacy policy. If ARCore is unavailable, EchoSense-AI may use camera-only estimates.

On compatible Apple devices, EchoSense-AI may use Apple camera, Vision, speech, translation, and depth-sensing frameworks. These system components are governed by Apple's applicable terms, privacy disclosures, and device settings.

Safety and depth information is experimental and must not be treated as a replacement for a cane, guide dog, sighted assistance, mobility training, or other established safety practices.

6. OCR, TRANSLATION, TEXT-TO-SPEECH, AND SYSTEM SERVICES

OCR, translation, text-to-speech, speech recognition, camera, audio, and accessibility services may be provided by the operating system, Google Play services, Apple, the device manufacturer, or another installed service. Some components may download language or model packs. Their processing and retention practices are controlled by their respective providers and your device settings.

7. ON-DEVICE MODEL DOWNLOADS

Large on-device language models are downloaded separately rather than bundled with EchoSense-AI. When you download a model, the hosting provider or content delivery network receives standard network information such as your IP address, request time, requested file, and user-agent information. Model hosts may include public endpoints operated by Hugging Face, Google, or their content-delivery providers. Their privacy policies apply to those requests.

EchoSense-AI stores downloaded models and download metadata on your device. You can delete downloaded models from Settings.

8. DIAGNOSTICS AND ANALYTICS

The operating system and application-distribution platform may collect crash, performance, installation, or basic diagnostic information according to your device and account settings. Builds of EchoSense-AI that include a crash-reporting or analytics service may send the diagnostics disclosed by that service.

EchoSense-AI is not designed to intentionally place camera images, document content, prompts, API keys, or complete model responses in diagnostic logs. However, you should review logs before sharing them because system frameworks and third-party components may produce their own diagnostic output.

9. DATA SHARING

EchoSense-AI does not sell personal information. Information is shared only as needed to provide a feature you request, such as with a selected online AI provider, model host, operating-system service, speech service, translation service, ARCore, crash-reporting service, or application-distribution platform.

Any third party with which EchoSense-AI shares user data is required to provide the same or equal protection of user data as described in this Privacy Policy and required by the Apple App Review Guidelines and applicable law.

EchoSense-AI may also disclose information when required by law or when reasonably necessary to protect users, the public, legal rights, or the security and integrity of the application.

10. DATA RETENTION AND USER CONTROLS

Most EchoSense-AI feature data is processed temporarily on the device. Downloaded models, preferences, securely stored online-provider credentials, and files that you explicitly save may remain until you delete them, remove the connection, clear the application's data, or uninstall the application.

Third-party services may retain information according to their own policies. To request access to or deletion of information held by an online AI provider or platform service, contact that provider directly and use its account controls.

11. CHILDREN'S PRIVACY

EchoSense-AI is not directed to children under the minimum age required to consent to data processing in their jurisdiction. Children should use the application only with the involvement and consent of a parent or legal guardian.

12. SECURITY

EchoSense-AI uses platform security features and secure credential storage where available. No storage or transmission method is completely secure. Do not scan, attach, or send highly sensitive documents, credentials, financial information, medical information, private keys, or confidential content unless you understand which local and online services are active and accept the associated risk.

13. INTERNATIONAL PROCESSING

Online providers, model hosts, platform services, and content-delivery networks may process information in countries other than the one where you live. Their privacy policies describe their international data-transfer practices.

14. CHANGES TO THIS POLICY

This Privacy Policy may be updated as EchoSense-AI features or service providers change. The effective date at the top of the policy indicates the latest revision. Material changes should be presented in the application or through the channel where the application is distributed.

15. CONTACT

For privacy questions or requests concerning EchoSense-AI, use the developer support contact provided on the EchoSense-AI application-store listing or the official EchoSense-AI distribution page. Questions about information processed by Apple, Google, OpenAI, Hugging Face, a device manufacturer, or another third-party service should be directed to that provider.