At TransVoice, we take your privacy seriously. This Privacy Policy explains how we collect, use, disclose, and safeguard your information when you use our voice training application. The data controller within the meaning of data protection laws is Kailar UG (haftungsbeschränkt), Viererblock 18a, 39326 Wolmirstedt, Germany.
Personal Information
When you create an account, we may collect:
- Name and email address
- Profile information you choose to provide
- Payment information (if you purchase supporter or premium features) - handled by Paddle, our merchant of record; we never receive or store your full payment card details
Voice Data
To provide our voice training services, we may collect:
- Microphone audio used for live analysis or replay, and recordings uploaded when you choose Saved mode or a sharing feature
- Voice analysis metrics calculated in your browser or, for a clearly indicated server-assisted feature, on our systems
- Practice session data and progress information
Usage Data
Using the Service necessarily sends technical request data such as your IP address, browser type, requested page and time of access to our systems. Additional audience measurement is sent to our self-hosted Umami service only after analytics consent. Umami uses the IP address, user agent and website ID to derive a pseudonymous session identifier and may also record the page, referrer, screen size, language and country. It sets no analytics cookies and does not track you across websites.
We process your personal data on the following legal bases under Article 6 GDPR:
- Contract Performance (Art. 6(1)(b) GDPR): For providing our services and managing your account
- Consent (Art. 6(1)(a) GDPR): For optional cloud storage of raw audio, separately enabled server-assisted AI features, optional improvement of voice-analysis models, community publication, audience measurement, and marketing communications. Where speech or voluntary text can reveal special-category data, we additionally request explicit consent under Article 9(2)(a) GDPR. Withdrawal takes future effect and does not automatically delete data already stored.
- Legitimate Interests (Art. 6(1)(f) GDPR): For fraud prevention, IT security, and necessary operational improvements based on technical data rather than voice content
- Legal Obligation (Art. 6(1)(c) GDPR): For retention of billing data as required by tax regulations
We use the information we collect to:
- Provide, maintain, and improve our services
- Personalize your voice training experience
- Track your progress and provide feedback
- Process transactions and send related information
- Send you updates, security alerts, and support messages
- Respond to your comments, questions, and requests
- Analyze usage patterns to improve the Service
- With separate explicit consent, use stored voice recordings and necessary alignment data to develop, train, test, and improve TransVoice voice-analysis and feedback models
- Protect against fraud and unauthorized access
Using your recordings to improve our models is optional and separate from every other TransVoice feature.
- Only after you consent may we use existing and future recordings in your account to improve voice analysis.
- Transcripts are used only to align events with the right place in the audio. They are not used as an independent text-training dataset.
- You can keep using TransVoice normally if you do not consent.
- You can withdraw at any time for future use. Your recordings will then be excluded from new selections and training runs.
- Consent does not allow speaker identification, voiceprints, voice clones, advertising, or selling your data.
What technically happens to voice data
Live pitch and loudness tools analyze the microphone signal in your browser. Microphone capture is required for that live analysis, but it does not by itself create a saved recording. In Private mode, short-lived audio used for feedback and replay remains in memory and is not added to your account journal. In Saved mode, audio is sent to our backend for analysis. Raw audio remains stored in DigitalOcean Spaces only when you enable raw-audio cloud storage; otherwise it is discarded after analysis. Derived metrics, classification results, curves and analysis artifacts may remain in your account and object storage in Saved mode independently of the raw audio. When you use AI chat, submitted audio is transcribed transiently by our self-operated Whisper service and the resulting text and necessary dialogue context are sent to the IONOS AI Model Hub; the transcription service does not permanently retain that audio. Dialogue text may be retained until you delete the associated training session or account. Voice-perception feedback and AI phrase generation use server processing only after the feature-specific consent shown in the app. With the separate voice-model-training consent, all recordings already stored in your account and recordings stored later may be used to develop, train, test, and improve TransVoice voice-analysis and feedback models. Associated transcripts are limited to aligning spoken events with time ranges in the audio and are not an independent text or language-model training corpus. Withdrawal excludes your recordings from later selection and training runs, but cannot undo a completed model run. This consent does not authorize speaker identification, voice cloning, sale, advertising, or a new external model provider. Community sharing remains a separate choice.
TransVoice is designed for trans and non-binary people and for anyone exploring their voice. We do not require you to state a gender identity or provide health information. Because speech and free text can still reveal sensitive details, the following boundaries apply:
- There is no required field for gender identity, transition status, diagnoses or medical treatment. Health information, details about sex life or sexual orientation, religion, political opinions, and information about another person are not required for any feature. Avoid adding such details to chat, custom phrases or other free-text fields unless you need them for practice content you chose yourself.
- The app asks separately for each optional purpose: voice-perception estimation, AI dialogue with transcription, phrase generation, and the use of stored voice data to improve voice-analysis models. Each decision is documented on the server with the notice version, language, time and any later withdrawal. A new notice version requires a new decision. None of these decisions enables cloud recording storage or community sharing.
- We analyze characteristics such as pitch, loudness and voice qualities for training feedback. We do not create a voiceprint, compare your voice to identify you, or use speaker-recognition technology.
- Creating an account or generally using the app is not treated as blanket consent under Article 9 GDPR. Before AI dialogue, phrase generation, or voice-model training, we expressly explain that voluntary speech or text may reveal special-category data and request explicit consent under Article 9(2)(a) GDPR. You may decline without losing the core service. Withdrawal stops future processing under that consent but does not automatically delete data already stored or undo completed model training; recordings and account data are removed through the available recording, session, or account deletion paths.
We do not sell your personal information. We may share your information with the following categories of recipients:
- Service Providers: With trusted third parties who assist us in operating our Service
- Legal Requirements: If required by law or in response to valid legal requests
- Business Transfers: In connection with a merger, acquisition, or sale of assets
- With Your Consent: When you explicitly agree to share information
Recipients and service providers
Depending on the features you use, the following recipients and service providers may process personal data:
- Google Ireland Limited and Google LLC (authentication via Google Sign-In, if selected by you) - Ireland and USA
- Apple Distribution International Limited and Apple Inc. (authentication via Apple Sign-In, if selected by you) - Ireland and USA
- Hetzner Online GmbH (hosting and infrastructure) - Germany (EU)
- DigitalOcean, LLC (cloud storage in the selected Frankfurt region for raw audio and analysis artifacts) - US company; content region Germany
- IONOS SE (AI-powered dialogue and practice-phrase generation via the IONOS AI Model Hub when you use the relevant feature) - Germany (EU)
- Self-hosted speech-to-text (Whisper) operated by Kailar UG on EU-based infrastructure - voice audio is transcribed transiently and is not retained by the service
- Paddle.com Market Ltd and affiliated Paddle companies (merchant of record for paid features: payment processing, invoicing, tax collection, fraud prevention and related legal obligations) - Paddle determines the purposes of the payment relationship and acts as an independent controller for it; processing outside the EEA may occur
- Plus Five Five, Inc., doing business as Resend (email delivery, particularly verification and transactional messages) - USA
- Self-hosted Umami web analytics operated by Kailar UG on Hetzner infrastructure in Germany (EU) - cookieless, pseudonymous audience measurement, no cross-site tracking; legal basis: your consent (Art. 6(1)(a) GDPR, Sec. 25(1) TDDDG), withdrawable at any time with future effect via cookie settings
Our core application infrastructure is located within the European Union. Some providers or affiliated companies are nevertheless established in the United States or may access data from there; an EU content region alone therefore does not eliminate a third-country transfer. For DigitalOcean, Google and Resend we rely, where the specific US entity remains validly certified, on the EU-US Data Privacy Framework and on contractually provided Standard Contractual Clauses as a fallback. Paddle.com Market Ltd is established in the United Kingdom; for transfers there we rely on the European Commission's adequacy decision for the United Kingdom, with Standard Contractual Clauses as a fallback. Apple describes transfers in particular under Standard Contractual Clauses. We regularly review the current mechanism and specific contract mapping. You can request a copy or further information at lea@transvoiceapp.com. Possible bases include:
- EU-US Data Privacy Framework for certified US companies
- Standard Contractual Clauses (SCCs) approved by the EU Commission
- Adequacy decisions by the EU Commission
As a data subject, you have the following rights:
- Right of Access (Art. 15 GDPR): You can request confirmation of whether we process your data and obtain information about this data
- Right to Rectification (Art. 16 GDPR): You can request correction of inaccurate data
- Right to Erasure (Art. 17 GDPR): You can request deletion of your data, unless legal retention requirements apply
- Right to Restriction (Art. 18 GDPR): You can request restriction of processing under certain circumstances
- Right to Data Portability (Art. 20 GDPR): You can receive your data in a structured, machine-readable format
- Right to Object (Art. 21 GDPR): You can object to processing based on legitimate interests
- Withdrawal of Consent: You can withdraw any given consent at any time with effect for the future
- Right to Lodge a Complaint: You have the right to lodge a complaint with a supervisory authority. The competent authority for us is the State Commissioner for Data Protection Saxony-Anhalt, Leiterstraße 9, 39104 Magdeburg, Germany.
To exercise your rights, contact us at lea@transvoiceapp.com
We retain your data only as long as necessary for the respective purposes:
- Account data: Until account deletion. Active-system records are erased during the account-deletion process; any remaining backup copy is retained only until scheduled overwrite under the backup cycle documented for that system and is not restored for ordinary operations.
- Raw audio: Only while cloud storage is enabled and until you delete the recording, session or account. Derived metrics, classification results and analysis artifacts from Saved training may remain independently of raw audio until the relevant deletion. Voice-model-training consent does not by itself extend storage of an account recording; withdrawal prevents later selection and training use but cannot reverse a model run already completed while consent was active.
- Operational and security logs: Limited to technical metadata and retained only as long as needed for troubleshooting, security and abuse prevention; overwritten or deleted under the documented log rotation and not used for retrospective content analysis.
- Billing records: For the retention period required by applicable tax and commercial law. TransVoice does not store full payment-card data.
- Support communications: Until the request is closed and afterwards only as long as needed for follow-up, legal defence or statutory obligations.
We do not use fully automated decision-making within the meaning of Art. 22 GDPR. The AI features used in our app for voice analysis serve solely to provide feedback and do not make decisions that have legal or similarly significant effects on you.
We use risk-based technical and organizational measures to protect your personal data. Data is transmitted over TLS/HTTPS. For stored data we use measures including access restrictions, separate service accounts, private-by-default sharing, deletion processes, and safeguards specific to each provider and system. We do not promise a particular server-side encryption mode for DigitalOcean Spaces objects, or blanket AES-256 encryption of all data stores, unless it has been technically verified for the actual storage path.
TransVoice is not intended for children under 16 years of age (in Germany). We do not knowingly collect personal information from children under 16. If we learn we have collected information from a child under 16, we will delete that information promptly.
We may update this Privacy Policy from time to time. For material changes, we will notify you by email or through a prominent notice in the app. The current version is always available on this page with the date of the last update.
If you have any questions about this Privacy Policy or our data practices, please contact us at lea@transvoiceapp.com