RemoteDataWork Privacy Policy

Version 1.0

Effective October 29, 2026

This Privacy Policy explains how Speechify, Inc. collects, uses, discloses and retains personal information when you visit remotedatawork.com, create or use an account, apply to RemoteDataWork, complete onboarding, take part in a recorded conversation project, receive payment, contact support or otherwise interact with the program.

RemoteDataWork is a paid-contributor program for adults in eligible United States jurisdictions and Canada outside Quebec. Contributors record natural conversations with one or more other enrolled Contributors to help train and evaluate speech AI. Recordings may also be included in datasets licensed to vetted organizations for their model training, but only as authorized through the separately signed Recording Release and Data License and the Contributor’s separate initials.

This Policy is not the Recording Release and Data License or the Voice Data Notice and Consent. Those documents are shown and signed during onboarding before any recorded session. The only audio recorded before signing is a short application voice sample, described in Section 3, which is used only to review your application and is deleted within 30 days after your application is decided. The Voice Data Notice and Consent provides a focused explanation of voice-data processing and obtains the required consent.

In summary:

• Speechify collects account, eligibility, identity-verification, consent, session, Recording, review, payment, support and technical information to operate RemoteDataWork.
• A Recording contains a person’s voice and remains personal information even after names and similar details are removed from its transcript.
• Speechify does not create or store voiceprints or speaker embeddings that identify or verify Contributors, and neither Speechify nor an authorized dataset recipient may create a replica of a Contributor’s voice.
• Every participant must separately tap to start every recorded session. Contributors separately initial use for Speechify model training and licensing to other organizations.
• A Contributor may request deletion of raw Recordings and exclusion from future dataset builds and model training runs. Datasets already delivered and models already trained are not recalled or retrained.

1. Who is responsible for personal information

Speechify, Inc., 382 NE 191st Street, PMB 69469, Miami, Florida 33179-3899, USA is responsible for the personal information described in this Policy. In this Policy, “Speechify,” “we,” “us” and “our” refer to Speechify, Inc.

Privacy questions and requests may be sent to privacy@speechify.com. Product support, appeals and deletion assistance are available at support@remotedatawork.com.

2. Scope

This Policy applies to RemoteDataWork websites, accounts, application and eligibility processes, contributor onboarding, Opportunities, recorded sessions, review and appeal processes, payments, support and related administration.

It does not replace the privacy notice of an identity, payment, tax, sign-in or other provider when that provider independently collects information from you. It also does not govern an authorized dataset recipient’s independent processing after a dataset has been delivered, although Speechify contractually restricts permitted recipients as described below.

3. Personal information we collect

Website and account information

We collect your name, email address, sign-in method, account and contributor identifiers, language preference, account status and authentication information. If you sign in through Google, Apple or another provider, we receive the account information that provider makes available with your permission. We use an authentication provider to operate and secure account sign-in.

We also collect IP address, browser and device information, approximate location derived from IP, dates and times, pages and features used, cookie and session identifiers, security events, errors and diagnostic information.

Application and eligibility information

We collect your state or province of residence, confirmation that you have reached the age of majority where you live, languages, project eligibility and the result of one-account-per-person and sanctions checks. If the program is not open where you live, we tell you at that step and collect no further application information.

Application voice sample. As part of your application, you may record a short voice sample (about two and a half minutes) after ticking a consent checkbox that is unchecked by default. We use it only to check your audio setup and language skills. It is never used to train AI models, added to a dataset, or used to identify you, including for duplicate-account checks, and it is deleted within 30 days after your application is decided, whether or not you are accepted. An application left incomplete for 30 days counts as decided.

The program does not use citizenship, national origin, race or ethnicity to determine eligibility, and we do not ask about race, ethnicity, national origin, citizenship, health information or similar sensitive traits. After you sign the contributor documents, we may ask optional, self-reported questions about your first language and other languages you speak, your accent or accent region, age range, occupation category and education level. You can skip or remove these answers, and they are never used to decide whether you can take part.

Identity verification information

An identity and eligibility provider may ask for a government-issued identity document or related information to confirm age, residence and that one person has one account. The current process does not require a selfie or face match. Speechify receives the verification result and related identifiers needed to administer the account. The provider may independently process the information you submit to it under its own notice.

Contributor documents and consent records

We collect signed copies and records associated with the Contributor Services Agreement, Recording Release and Data License, Voice Data Notice and Consent, W-9 or W-8BEN, Conversation Guidelines and this Privacy Policy. These records include the document version and language, date and time, contributor identifier, electronic signature, acknowledgements, separate initials for Speechify training and outside dataset licensing, and the tap-to-start confirmation for each session.

Tax forms may contain an address, date of birth, Social Security number, taxpayer identification number, foreign tax identifier or other information required by tax law. We restrict access to tax information and use it only for payment, reporting, compliance, fraud prevention and related legal purposes.

Recordings, transcripts and session information

We collect the audio of each recorded session, transcripts, the topic, date, duration, participant identifiers, consent state, recording indicator state, technical and audio-quality information, and information needed to match participants. The other participants hear what you say during the session.

We may identify and remove or mask names, telephone numbers, addresses, account numbers and similar personal details in transcripts and audio. We call the resulting material “PII-redacted,” not anonymous. A voice recording may still identify or be linked to a person.

We do not create, store or deliver a voiceprint or speaker embedding that identifies or verifies who is speaking. We do not use identity voiceprints for sign-in, duplicate-account detection, fraud detection, quality review, diarization or another program function.

Qualification, quality review and appeals

We collect setup-test results, qualification results, acceptance or rejection status, review notes, rejection reasons, appeal submissions, appeal outcomes and communications about the decision. Automated tools may assist with technical quality, synthetic-audio or policy checks, but a rejection is accompanied by a reason and may be appealed through the account.

Safety and conduct information

If you report another person, or another person reports you, we collect the report, relevant session and account information, communications, review notes and action taken. We may preserve relevant material while investigating a safety, fraud or legal issue.

Payment and tax information

We collect the Opportunity rate, accepted recorded time, qualification fee, amounts earned and paid, payment date and status, payout method and limited payment-account information. PayPal, Venmo, a banking provider or another payment provider may collect the financial information needed to send payment. Speechify receives transaction identifiers and status information needed to reconcile payment and meet tax and accounting obligations.

Support and communications

We collect messages sent to support@remotedatawork.com or another Speechify address, including account, appeal, deletion and privacy requests, attachments, verification information and records of our response.

4. Sources of personal information

We obtain personal information:

• directly from you through the website, account, documents, sessions, reports and communications;
• from other participants in a matched session;
• automatically from your browser, device and use of the Platform;
• from sign-in, identity, payment, tax, hosting, communications and security providers; and
• from Speechify records created when we review a Recording, process an appeal, make a payment, investigate a report or build and track a dataset.

5. How we use personal information

We use personal information to:

• create, authenticate, administer and protect accounts;
• determine age, residence, location and project eligibility and enforce one account per person;
• provide and document the required agreements, notices, signatures, initials, acknowledgements and session consent;
• match eligible Contributors and operate recorded sessions;
• review application voice samples and conduct setup and qualification sessions;
• transcribe and PII-redact Recordings;
• review Recordings under the published Acceptance Criteria, explain decisions and decide appeals;
• calculate and send payments and complete tax reporting;
• provide support, investigate reports and enforce safety and conduct rules;
• detect fraud, synthetic or replayed audio, account misuse and security incidents without creating an identity voiceprint;
• create datasets, document their provenance and track which Recordings are included in an internal training run or external delivery;
• train and evaluate Speechify speech and dialogue models when you have separately initialed that use;
• license datasets to vetted organizations for their model training when you have separately initialed that use;
• research, test and improve Speechify’s products and services that use speech;
• honor deletion requests by deleting raw Recordings under Speechify’s control and excluding them from future dataset builds and model training runs;
• comply with tax, accounting, sanctions, privacy and other legal obligations and establish, exercise or defend legal claims; and
• create aggregated statistics that do not reasonably identify a person.

Speechify does not use a Contributor’s name or voice to endorse a product, service, political position or message. Speechify does not use a Recording to create a replica of the Contributor’s voice. These restrictions also apply to authorized dataset recipients through the Buyer Data License Agreement.

6. Consent and other authority for processing

We ask for express, specific agreement before collecting and using Recordings for the program. During onboarding, you separately initial use for training and evaluating Speechify models and licensing datasets to other organizations for model training. Before each session, every participant receives a recording and AI-training notice and must tap to start before recording begins.

We process other information as necessary to administer the agreements you enter into, operate the Platform, review and pay for contributions, meet legal obligations, protect people and systems, and pursue legitimate operational and legal interests. In Canada, we obtain meaningful consent where required and provide the choices and information required by applicable federal and provincial privacy law.

After an accepted Recording has been paid, the Recording Release and Data License is irrevocable to the extent stated in that document. You may still request deletion and future exclusion as described in Section 11. Withdrawing consent or closing an account does not affect processing that was lawful before withdrawal, recall a delivered dataset or require retraining of an existing model.

7. How we disclose personal information

We disclose personal information only as described in this Policy, the signed contributor documents, at your direction or as permitted by law.

Other participants

The other participants in a session hear your voice and statements and may see the participant information displayed in the session interface. Do not disclose information you do not want them to hear.

Service providers

We use providers for authentication, identity and eligibility checks, hosting and storage, transcription and redaction, security, email and support, payments and tax administration. Providers receive only the information reasonably needed for their function and are required to protect it and use it only for authorized purposes. Where applicable law requires it, you may request a list of specific third parties to which we have disclosed personal information.

Speechify personnel and contractors

Authorized personnel and contractors may access application, Recording, review, payment, support and technical information when needed to operate and secure the program, review quality, decide appeals, provide support, create datasets, investigate a report or meet legal obligations. Access is limited by role and operational need.

Dataset recipients

Speechify may license datasets to vetted organizations only where the Contributor has separately initialed that use. A delivered dataset may include audio, PII-redacted transcripts, Session Information (including optional labels), rotated speaker identifiers, technical information and a dataset manifest. Names, contact details, account identifiers and tax or payment information are not included.

Before any delivery, Speechify conducts buyer diligence and requires a Buyer Data License Agreement. The recipient may use the dataset only to train and evaluate speech and conversation models. The recipient may not create a voice replica, a voiceprint or a system that identifies speakers; may not try to identify any Contributor; may not sell, sublicense or publish the dataset, except that it may share it with its affiliates and service providers bound by the same terms; and must protect the dataset with reasonable security.

Speechify does not license datasets containing Recordings to anyone located in, or owned or controlled from, a country subject to comprehensive U.S. sanctions or designated by the U.S. government as a country of concern for the protection of Americans' sensitive personal data. Buyer diligence considers ownership and control, not only the contracting address.

Professional advisers, authorities and transactions

We may disclose information to lawyers, auditors, insurers and other advisers subject to appropriate duties. We may disclose information when reasonably necessary to comply with law or valid legal process, investigate fraud or misuse, protect rights or safety, or establish, exercise or defend legal claims. Information may be transferred as part of a merger, financing, acquisition, reorganization, bankruptcy or sale of relevant assets, subject to applicable law and the contributor documents.

8. Dataset use and important limits

Recordings are used first for Speechify model training and evaluation and may later be included in datasets licensed to authorized recipients, in each case only under the uses separately initialed in the Recording Release and Data License.

Speechify records the consent version, jurisdiction and deletion status associated with each Recording and tracks which Recordings enter each internal training run and buyer delivery. Dataset manifests record relevant counts, document versions, jurisdictions and date ranges. Speaker identifiers are rotated between recipients so that different recipients cannot use a common identifier to link the same Contributor.

A deletion request prevents a raw Recording from being used in any dataset build or model training run that begins after the request is processed. It does not retrieve a dataset already delivered to a recipient. It also does not remove information already learned by or require retraining of a model trained before the request. Speechify retains the records needed to document consent, provenance, delivery, payment and the deletion request.

9. Cookies and similar technologies

RemoteDataWork uses cookies, local storage and similar technologies for sign-in, account security, session continuity, preferences and core website functions. We may also use limited analytics, fraud-prevention and performance technologies to understand use and protect the Platform.

Where applicable law requires consent for a non-essential technology, we will request it before activation. You can use available cookie controls and browser settings to manage those technologies, although blocking necessary cookies may prevent account or session functions from working.

We do not use RemoteDataWork personal information for third-party targeted advertising and do not sell or share website browsing data for cross-context behavioral advertising. Licensing contribution datasets for the separately authorized training purposes described in this Policy is different from advertising and is disclosed directly in the signed contributor documents.

10. Retention

We retain personal information only as long as reasonably necessary for the program, the purposes described in this Policy and applicable legal obligations. How long we keep application voice samples, Recordings, transcripts, Session Information, signed contributor documents, consent records, session start records and payment records is set by Section 5 of the Voice Data Notice and Consent and by our Contributor Voice Data Retention Policy at remotedatawork.com/legal/retention. If this Policy differs from those documents on those records, those documents control. For other information, the current schedule is:

Incomplete applications. Generally up to 12 months after the last application activity, unless the information is needed for security, fraud prevention or a legal obligation. Application voice samples are deleted sooner, as described in the Retention Policy.

Unsuccessful applications. Generally up to two years after the final decision, so that Speechify can administer appeals, prevent duplicate or abusive applications and document eligibility decisions. Information may be kept longer for an unresolved fraud, safety or legal matter. Application voice samples are deleted sooner, as described in the Retention Policy.

Account, profile and eligibility information. While the account is active and generally for two years after closure. A limited eligibility, sanctions, duplicate-account or deletion record may be retained longer where necessary to enforce a restriction or legal obligation.

Identity-verification result. While the account is active and for up to two years afterward. The identity provider’s retention of underlying documents is governed by its notice and its contract with Speechify.

Review, rejection and appeal records. Generally three years after the final decision, and longer if needed for an unresolved payment, safety, fraud or legal matter.

Dataset manifests, lineage and deletion flags. For as long as the related internal model or delivered dataset remains in use or otherwise relevant, and for the applicable claims period afterward. These records document authorization, provenance, delivery scope and deletion status and may be retained after the underlying raw Recording is deleted.

Tax forms and other tax and accounting records. For the period required by applicable tax and accounting law and Speechify's documented financial-records schedule, and at least as long as the Voice Data Notice and Consent requires for payment records.

Safety, fraud, support and privacy-request records. Generally three years after the matter closes, and longer where reasonably necessary for an investigation, legal claim or regulatory obligation.

Standard website and security logs. Website logs are generally kept for 90 days and security or access logs for up to one year, unless a specific incident or legal hold requires longer retention.

Backups. Removed or overwritten on normal cycles, generally within 90 days after deletion from active systems, unless isolated for security or a legal hold.

Aggregated information that no longer reasonably identifies a person may be retained longer. A delivered dataset is controlled by the recipient under its Buyer Data License Agreement and is not recalled when a Contributor later deletes an account or Recording.

11. Deletion and withdrawal

You may request account and data deletion through the account page or by emailing support@remotedatawork.com from the address associated with the account. We may verify identity before acting.

After a verified request, Speechify will:

• close or restrict the account as appropriate;
• delete raw Recordings and linked transcripts under Speechify’s control, subject to a specific legal hold or other lawful exception;
• mark those Recordings so they are excluded from all future dataset builds and model-training runs;
• delete or de-identify account and application information that is no longer needed; and
• retain only the consent, provenance, lineage, payment, tax, safety, fraud, legal and technical records described in Section 10.

Deletion does not recall a dataset already delivered, delete a recipient’s authorized copy or require Speechify or a recipient to retrain or alter a model already trained. If information must be retained for an existing or reasonably anticipated legal claim, investigation, regulatory matter, security incident, tax obligation or other legal requirement, Speechify will restrict it to that purpose and delete it when the reason ends.

In Canada, you may withdraw consent subject to legal or contractual restrictions and reasonable notice. We will explain the consequences. Withdrawal may end eligibility to participate and does not undo a use or disclosure that occurred lawfully before withdrawal.

12. Security

Speechify uses administrative, technical and organizational measures designed to protect personal information. These include encryption in transit and at rest where appropriate, role-based access, named access to Recordings and tax information, access logging, authentication, provider controls and incident response procedures.

No system is completely secure. Protect your account credentials, use the minimum personal information needed in a conversation and contact support@remotedatawork.com promptly if you suspect unauthorized access.

13. International and cross-border processing

Speechify is established in the United States. Personal information may be stored and processed in the United States and in other countries where Speechify or its providers operate. Canadian personal information may therefore be accessible to courts, law-enforcement agencies or other authorities under the laws of those countries.

Speechify uses contracts and other safeguards required by applicable law when transferring personal information. Dataset recipients are limited and vetted as described in Section 7. RemoteDataWork does not accept Contributors located outside the eligible United States and Canadian jurisdictions described in the Terms.

14. Your privacy rights

Depending on where you live and subject to legal exceptions, you may have the right to request access to personal information, correction, deletion, a copy in a portable format, information about disclosures or specific third parties that received personal information, restriction of certain uses, withdrawal of consent or an appeal of a denied privacy request. You may also challenge Speechify’s compliance and complain to an applicable privacy regulator.

Submit a request through the account page or email privacy@speechify.com. Identify the account and the right you wish to exercise. We may request information reasonably necessary to verify identity and authority. We will respond within the period required by applicable law and will explain a denial and any available appeal route where required.

An authorized agent may submit a request where the law permits. We may require proof of authority and may verify your identity directly. Speechify will not discriminate against you for exercising a privacy right.

Canada

Canadian Contributors may request access to and correction of personal information under Speechify’s control, withdraw consent subject to legal and contractual limits, and challenge compliance. Speechify remains accountable for information processed by providers outside Canada. You may complain to the Office of the Privacy Commissioner of Canada or, where applicable, the privacy commissioner in Alberta or British Columbia.

United States

Residents of a United States state with an applicable privacy law may have rights to know, access, correct, delete or obtain a copy of personal information and to appeal certain decisions. RemoteDataWork does not sell or share personal information for targeted advertising. The commercial licensing of contribution datasets occurs only under the separately signed and initialed contributor documents and is described in Sections 7 and 8.

15. Notice at collection for United States residents

During the preceding 12 months, Speechify may have collected the following categories of personal information for the purposes described in Section 5:

• identifiers and contact information;
• account, application, eligibility and identity-verification information;
• internet, device, cookie, usage and coarse location information;
• audio information, transcripts and other session information;
• professional, project, payment, tax and transaction information;
• communications, support, report and appeal information;
• optional first language, accent, age range, occupation category and education level information; and
• inferences and decisions associated with eligibility, quality, safety, fraud and appeals.

We collect these categories from the sources described in Section 4, disclose them to the recipients described in Section 7 and retain them under Section 10. Voice recordings and account credentials may be treated as sensitive personal information under some laws. Speechify uses sensitive information only for the disclosed program, security, payment, compliance and authorized training and dataset purposes, not to infer unrelated characteristics.

16. Adults only

RemoteDataWork is for adults who satisfy the eligibility requirements. We do not knowingly permit a minor to create a contributor account or take part in a session. No minor or unenrolled person may be audible in a Recording. Contact privacy@speechify.com if you believe a minor’s information was submitted so that we can investigate and delete or restrict it as required.

17. Changes to this Policy

We may update this Policy as the program, providers or law changes. The version in force and its effective date appear at the top. If a change materially affects an enrolled Contributor or an authorized use of Recordings, we will provide additional notice and obtain a new acknowledgement, signature or consent where required. A revised Policy does not retroactively expand the authorized use of a Recording.

18. Contact

Privacy questions and requests: privacy@speechify.com

Program support, appeals and deletion assistance: support@remotedatawork.com

Speechify, Inc.

382 NE 191st Street, PMB 69469

Miami, Florida 33179-3899

USA