This policy describes what this service collects, why, where it is kept and what you can do about it. It is written against what the software actually does.
Who is responsible
This service is operated by Ishaan Verma, trading as Elsewhere. For anything in this policy, including requests about your data, write to contact@listenelsewhere.com, or by post to Flat No. A 01/404, Tower II, Plot No. 93/1, Silver City, Sector 93, Noida, Uttar Pradesh 201304, India.
What is stored on our server
When you sign in and add a book, the following is kept in our database:
- Your account. The name, email address and profile picture URL that Google returns when you sign in, and the sign-in session itself.
- Your books, as text. The full text of every book you add, broken into chapters and passages, along with titles, authors and chapter structure. The text of your books is uploaded to and stored on our server.
- Your books’ images. The cover and any images inside the book file are stored as files in our file storage, the same way the audio described below is when you choose the cloud for it. That storage is our server in the European Union or Cloudflare R2 in the European Union; see “Who else can see it” below.
- Where you are in each book — the passage and the offset within it — so that reading resumes on your next visit.
- Your highlights and notes, as the passage and character range they cover plus any note you wrote.
- Your settings, such as the chosen voice and playback speed.
- Your plan. If you become a Member, what Paddle tells us about the subscription: its identifiers, whether it is active, monthly or yearly, when the paid period ends, and when a payment first failed, if one did. We never receive your card number or other payment details.
- A record of which passages have narration, its duration and which voice produced it. The audio itself is stored on our server only if you choose the cloud for that book; see below.
Where your narration audio lives
You choose, per book, where its narration audio is kept:
- In your browser — the default. Generated audio is written to your browser’s storage on the machine that made it. It is not uploaded, it does not sync between your devices, and clearing your browser’s site data deletes it.
- In our cloud storage, if you move a book to the cloud instead. Its audio is then stored in our file storage, on our server in the European Union (rented from Contabo) or in Cloudflare R2 in the European Union, up to 1 GB per account on the free plan and more for Members. It is deleted when you delete the book, delete your account, or move that book’s audio back to a device. If a Member subscription ends, books beyond your 5 newest, with their text and images, and then the cloud audio of kept books above 1 GB, can also be removed after 14 days, as the terms describe. Audio on your devices for the books you keep is not touched; a removed book’s audio is cleared from a device the next time it opens your library, so export it first as an M4B from the book’s Audio Studio. You can also delete a book’s cloud audio on its own, keeping the book, from the cloud storage list on your settings page.
Backing up audio yourself. From the Audio Studio on a book’s page, you can save a backup of that book’s audio to your own device. That file is yours to keep; we never receive it. Restoring one is handled entirely by your browser, which reads the file and writes its contents straight into its own storage — we only learn which device now holds that book’s audio, never the file itself.
The voice model and the generation itself. Narration is produced by a model that runs in your browser. The text of your book is not sent to any third-party speech service to make it.
What we do not do
- We do not sell or share your personal information, in any sense those words carry under the CCPA or CPRA.
- We do not serve advertising, and we run no advertising or analytics trackers. Error monitoring, described below, is the only third-party code that ever reports anything about your use of the service, and it reports faults rather than behaviour.
- We do not use the text of your books to train any model.
- We do not read your books, except where we must look at a specific record to investigate a fault you have reported or to respond to a legal demand.
Cookies
We set a cookie to keep you signed in, and store a small preference locally for your reading theme. There are no advertising, analytics or third-party tracking cookies, which is why you are not asked to consent to any.
Who else can see it
Google handles sign-in, and knows that you signed in to this service. Their handling is governed by Google’s own privacy policy. The service runs on a server rented from Contabo, in the European Union, and Cloudflare sits in front of it for DNS, TLS and caching. Both necessarily process data in transit, and Contabo at rest, on our behalf.
Paddle (Paddle.com) is the merchant of record for every purchase, and takes your payment. When you buy Member you give your payment details and billing country to Paddle, not to us. We pass Paddle your account’s email address and an internal account identifier, so the payment is linked to your account. Paddle handles the payment, the tax and the invoice, and we never see your card. Paddle tells us the state of your subscription so the service can apply the right limits. Paddle’s handling of your data is governed by its own privacy policy.
Resend sends email on our behalf: the feedback you submit from the app, which reaches us with your email address as the reply address, and notices about your membership. It receives the address and the message.
Cloudflare R2 is an object store, run by Cloudflare, that holds your books’ images and the audio you choose to keep in the cloud, in place of the storage on our own server. Our R2 storage is in the European Union, as is our server. Those files are stored by Cloudflare on our behalf, and are deleted from it when they are deleted from the service.
Hugging Face hosts the speech model your browser downloads the first time you generate narration. Your browser fetches it from them directly, which tells them that a device is downloading the model. No part of your book is involved in that request, and the narration itself is produced on your machine.
Sentry receives error reports, from your browser and from our server, when the operator of this instance has configured it. It is off unless a reporting address is set, so an instance someone runs for themselves has no such processor at all. A report carries the address of the page or request that failed (which includes the identifier of the book you were on), the opaque identifier of your account, the recent steps that led to the fault, and — because it is an ordinary network request — your IP address, which we do not ask them to record but cannot prevent them receiving. The text of your books is deliberately removed before anything is sent: request bodies are dropped whole rather than filtered by field, and the local variables attached to each line of a stack trace are stripped, because a passage being processed when something failed would otherwise be sitting in one.
We will disclose data where we are legally required to, and we will tell you when we are permitted to tell you.
How long it is kept
Your books, positions and highlights are kept until you delete them or delete your account, or until a lapsed Member subscription runs out its 14 days and the books beyond your 5 newest are removed, as the terms describe. Deleting your account removes your account record, your books and their text and images, your positions and your highlights from our database and our file storage, including any audio you had moved to the cloud, and cancels any Member subscription. Paddle keeps its own records of your purchases, as it is required to by law. Audio already on your own device is yours to keep or clear; deleting your account does not reach into your browser storage.
One thing outlives a deletion, briefly. If a free account had narrated anything that month, we keep a one-way hash of your Google account ID and that month’s narration total until the month ends, so that deleting an account and signing up again does not reset the monthly allowance. It cannot be turned back into your ID or your email, and it is removed when the month turns over.
Your rights over your data
Whoever and wherever you are, you can do both of these from your settings page, without asking us:
- Export everything we hold about you, as a file. The file holds your account, book text, positions and highlights but not audio; audio can be saved from each book’s Audio Studio, as an M4B export or an audio backup.
- Delete your account and its contents.
If you are in the UK, the EEA or Switzerland, the GDPR gives you rights of access, rectification, erasure, restriction, portability and objection, and a right to complain to your data protection authority. If you are in California, the CCPA and CPRA give you rights to know, delete, correct and limit, and a right not to be discriminated against for exercising them. Since we neither sell nor share personal information, there is nothing to opt out of, and Global Privacy Control signals accordingly change nothing about how we treat you. To exercise anything the two buttons above do not cover, write to contact@listenelsewhere.com.
Security
Access to your books requires your session, and every request for a book, chapter, passage, highlight or image is checked against the account that owns it. No transmission or storage is perfectly secure, and we do not claim otherwise.
Children
This service is not intended for children under 13, and we do not knowingly collect their information. If you believe a child has given us data, write to us and we will delete it.
Changes
If this policy changes materially we will say so on this page and update the date at the top before the change takes effect.