PRIVACY POLICY — Voice Typer

WHO WE ARE

Voice Typer is developed and operated by Estri Tech, an individual developer based in India. When this policy says "we", "us", or "our", it refers to Estri Tech.

Developer: Estri Tech
Contact: estritech@gmail.com
Address: New Delhi, India

1. INFORMATION WE COLLECT

Account Information

When you register or sign in, we collect your email address. You can register with an email and password, or sign in with your Google account. We do not have access to your password — passwords are managed entirely by our authentication provider and are never stored or readable by us.

Note: if you register with an email and password rather than Google Sign-In, we do not currently offer a self-service password reset. If you lose access to your account, signing in with Google using the same email address is the recovery path.

Device Identifier

When Voice Typer first runs, it generates a unique identifier tied to your physical device. This identifier is hashed before it is ever transmitted or stored — the raw value never leaves your machine. We use it solely to prevent repeated abuse of the free trial and to enforce the two-device maximum for Pro subscribers.

Account and Session Records

We store basic account metadata in our database: when your account was created, when you last signed in, the hashed device identifiers associated with your account, and which devices are currently active under your subscription. This is used exclusively for account management and license enforcement.

Usage and Analytics Data

We collect anonymised usage events to understand how the app is used and to guide product improvements. This is handled through PostHog, an analytics platform whose EU servers store this data. We never capture the content of your speech or transcriptions. The events we collect are:

• Dictation session metadata: The application you were dictating into (e.g. a browser, a word processor, a text editor); the language profile you had selected; which speech engine processed the audio; how long the dictation session lasted in seconds.
• Application lifecycle and UI interactions: When the app is launched, minimised, or closed; which settings tabs you navigate to; changes you make to app preferences (e.g., toggling a feature on or off, or changing the interface language); trial and upgrade journey events (when a free trial starts, when it expires, and when upgrade-related buttons are clicked).

You may opt out of analytics collection at any time in the app's settings. Opting out does not affect any core functionality.

Failed-Transcription Audio Clips

To improve transcription accuracy, we collect short audio recordings from sessions where speech activity was detected but no transcription was produced. We do not collect audio from successful transcription sessions, and we do not collect recordings where only silence or background noise below our detection threshold was present.

This collection is enabled by default. We highlight it clearly during the onboarding flow so you are aware before it first occurs. You can disable it at any time in the Account tab of the application.

Each clip is submitted together with the following metadata:

We do not receive or store a readable transcript alongside these clips, because none was produced. Clips are used solely to diagnose and improve transcription accuracy. They are not used for any other purpose, not shared with third parties for their own use, and not used to train third-party models.

Payment Information

If you purchase a Pro subscription, payment is processed by DodoPayments, a third-party payment processor based in India. They collect and store your name, email address, phone number, billing address, and card details. We do not store card details on our own servers. As the merchant, we can view your name, email, phone number, and address in the DodoPayments console for the purpose of resolving billing disputes and managing your subscription.

What We Do Not Collect on Our Servers

• Your transcription history and dictated text.
• Audio from successful transcription sessions.
• Silent recordings — clips are only submitted when speech activity is detected and no transcription is returned.
• Your custom word substitutions and shorthand entries — these live in a local file on your device only.
• Your debug log — this is stored locally and is never automatically uploaded. If you contact us for support and choose to share it, we use it only to resolve your issue.
• Your app settings — all preferences are stored locally on your machine.

2. HOW YOUR AUDIO IS PROCESSED

Standard Engine (Free, Unlimited): When you use the Standard Engine, your audio streams directly from your device to Google's Speech-to-Text API. Google's handling of your audio is governed by Google's own Privacy Policy. Along with that, simultaneously, a request is also made to DeepInfra's server to use Whisper V3 Large Turbo to get transcription. This audio, even though it goes through our server, is never stored.

Chirp Pro Engine (Pro Subscription): When you use the Chirp Pro engine, your audio is routed through a proxy server we operate on Microsoft Azure infrastructure in the South Asia region. The proxy forwards your audio to Google Cloud Speech-to-Text, receives the transcription, and sends it back to your device. Audio is processed entirely in memory during your active session and is never written to disk. Transcription content is not logged. Server-side logs capture only connection metadata such as timestamps, session duration, and error codes — nothing from the content of your speech is retained after your session ends.

Failed-Transcription Audio Collection: When speech activity is detected but no transcription output is produced, and you have not opted out of audio improvement collection, the audio clip is transmitted to our proxy server over an encrypted connection and stored for quality improvement purposes. Clips are retained for no longer than 30 days and are deleted on a rolling basis after that period. Metadata associated with these clips may be retained in anonymised aggregate form for longer for trend analysis purposes. Clips are not shared with third parties and are not used to train third-party models.

3. DATA STORED LOCALLY ON YOUR DEVICE

Voice Typer stores a number of items locally on your Windows machine. These include:

• Registry entries used to store your trial and subscription status, a fallback device identifier if hardware reads are unavailable, and a startup entry if you enable that feature in settings.
• A configuration file at a standard application data path, containing your preferences, language profiles, hotkey assignments, word substitutions, subscription status, and your audio improvement opt-out preference.
• A debug log file that may include diagnostic information and transcription output from your sessions.

You may delete any of these files or registry entries at any time without affecting your cloud account.

4. WHY WE COLLECT YOUR INFORMATION

We collect your email address and session records to create and manage your account and deliver the service. We collect a hashed device identifier to prevent trial abuse and enforce device limits, which are legitimate business interests. Payment data is processed to fulfil your subscription contract. Usage analytics are collected on the basis of legitimate interests in improving the product — you may opt out at any time. Failed-transcription audio clips are collected on the basis of your consent — you are informed during onboarding and may opt out at any time via the Account tab. Promotional emails are sent only with your consent, which you may withdraw at any time.

For users in the European Economic Area or United Kingdom, the specific legal bases are:

• Performance of contract: account data and payment processing.
• Legitimate interests: device identifier collection for fraud prevention; usage analytics for product improvement (with opt-out available).
• Consent: failed-transcription audio clips (highlighted during onboarding; opt-out available at any time in Account settings); promotional email communications.

5. THIRD-PARTY SERVICES

We use the following third-party services. Each operates under its own privacy policy:

• Google Speech-to-Text: powers the Standard Engine. Your audio streams directly to Google from your device and does not pass through our infrastructure.
• DeepInfra: powers the Whisper V3 Large Turbo model for high-accuracy simultaneous transcription. Audio routed through DeepInfra is processed in real time and is never stored on our servers.
• Google Firebase: handles account sign-in and stores your email address, hashed device identifier, and session timestamps. Firebase infrastructure is operated by Google in the United States.
• Google Cloud Speech-to-Text v2: powers the Chirp Pro engine via our proxy server.
• Microsoft Azure: hosts our proxy server in the South Asia region. In-session Chirp Pro audio is routed through this infrastructure and is never written to disk.
• Google Firebase Storage: stores failed-transcription audio clips submitted for quality improvement purposes. Firebase Storage is operated by Google and subject to Google's privacy and security standards.
• DodoPayments: is our payment processor, based in India. They handle all billing details for Pro subscriptions.
• PostHog: is our product analytics platform. Usage event data is stored on PostHog's EU-region servers. PostHog does not receive any audio or transcription content.
• Brevo: delivers our promotional emails. Only your email address is shared with Brevo for this purpose. Brevo is based in France.

We do not use Facebook Pixel, Google Analytics, advertising networks, or any behavioral retargeting tools.

6. INTERNATIONAL DATA TRANSFERS

Estri Tech is based in India. Your data may be processed in other countries depending on which services are involved. Firebase and Google Cloud operate primarily in the United States. Our proxy server and audio storage infrastructure operates in the South Asia region on Microsoft Azure. DodoPayments operates in India. PostHog analytics data is stored in the European Union. Brevo operates in France. Where we rely on third-party processors, we rely on their respective compliance frameworks for any cross-border transfers.

7. DATA RETENTION

We retain your account record — including your email address, session timestamps, and hashed device identifier — until you request deletion. Your active device list is retained until you deactivate devices or your subscription ends. Analytics event data is retained in PostHog subject to PostHog's data retention settings and our own configuration. Payment records are retained by DodoPayments in accordance with their policy and applicable financial regulations. Proxy server session metadata is subject to standard server log rotation. Failed-transcription audio clips are retained for no longer than 30 days following submission, after which they are deleted on a rolling basis. Anonymised aggregate metadata derived from these clips may be retained for longer for trend analysis. Local files on your device remain until you delete them.

To request deletion of your account data, including any audio clips associated with your device identifier, contact us at estritech@gmail.com. We will process your request within 30 days.

8. YOUR PRIVACY RIGHTS

All Users: Regardless of where you are located, you may request access to the personal data we hold about you, request correction of inaccurate data, request deletion of your account and associated data (including audio clips linked to your device identifier), opt out of analytics data collection via the app settings, opt out of failed-transcription audio collection via the Account tab, and unsubscribe from promotional emails at any time. Contact us at estritech@gmail.com for any of these requests.

EEA and UK Residents (GDPR and UK GDPR): If you are located in the European Economic Area or United Kingdom, you have the right to access a copy of your personal data, request correction of inaccurate or incomplete data, request erasure of your data where we no longer have a lawful reason to retain it, request restriction of processing, receive your data in a portable format, and object to processing. To exercise any of these rights, contact us at estritech@gmail.com. We will respond within 30 days.

California Residents (CalOPPA and CCPA/CPRA): We do not sell your personal information and we do not share it for cross-context behavioral advertising. Audio recordings fall within the category of sensitive personal information under the CPRA — we collect them solely for transcription quality diagnostics, do not sell them, and provide an instant in-app opt-out. To make a California privacy request, contact estritech@gmail.com.

9. CHILDREN'S PRIVACY

Voice Typer is a general-purpose productivity application and is not directed at children. We do not knowingly collect personal information from children under the age of 13, or under 16 where required by applicable law including users in the European Economic Area. If you are a parent or guardian and believe your child has provided us with personal information, please contact us at estritech@gmail.com and we will delete it promptly.

10. PROMOTIONAL EMAILS

With your consent, we may send you promotional emails about Voice Typer updates, new features, and offers. These are delivered via Brevo. Every promotional email will contain a clearly visible unsubscribe link. You may also opt out at any time by emailing estritech@gmail.com with the subject "Unsubscribe".

11. ANALYTICS AND AUDIO COLLECTION OPT-OUT

You can disable analytics event collection at any time via the app settings, or by contacting us at estritech@gmail.com with your account details.

You can disable failed-transcription audio collection at any time via the Account tab in the application. Opting out stops all future clip submissions from your device. To request deletion of clips already submitted, contact us at estritech@gmail.com with your account email and the subject "Audio Deletion Request" — we will delete them within 30 days. Neither opt-out affects any features or the quality of the service you receive.

12. DO NOT TRACK

Voice Typer is a desktop application and does not currently respond to Do Not Track signals from browsers. We do not engage in cross-site tracking or online behavioural advertising.

13. SECURITY

We take reasonable steps to protect your personal data. Device identifiers are hashed before being stored on our servers. All data in transit uses encrypted connections, including audio clips submitted for quality improvement. Audio processed by the Chirp Pro engine during active sessions is handled in memory only and is never written to disk. Failed-transcription audio clips are stored on Google Firebase Storage and are subject to Google's security controls and infrastructure standards. Analytics data is stored in the EU by PostHog.

14. CHANGES TO THIS POLICY

We may update this privacy policy from time to time. When we do, we will update the date at the top of this document. For material changes that meaningfully affect how we handle your data, we will notify you by email if you have an account with us, or by a notice within the application. Continued use of Voice Typer after changes take effect constitutes your acceptance of the revised policy.

15. HOW TO CONTACT US

For any privacy-related questions, requests, or complaints, contact us at:
Estri Tech
New Delhi, India
Email: estritech@gmail.com