Privacy Policy
- Application: ToneSphere
- Publisher: Neural Nexus Studios, an individual developer, Kolkata, West Bengal, India
- Policy version: 1.3
- Effective date: 2 October 2026
1. Summary
ToneSphere collects no user information.
There are no user accounts, no sign-in, no registration, no telemetry, no analytics, no advertising identifiers, no crash or usage reporting, and no third-party tracking of any kind. When something goes wrong, ToneSphere writes a crash report to a file on your own computer and tells you where it is; it never sends it anywhere. ToneSphere is a desktop audio router: it takes audio from the devices and applications on your computer, processes it, and sends it to other devices on the same computer. That audio is handled in memory, in real time, and is neither stored nor transmitted by ToneSphere.
There is exactly one feature that sends audio away from your machine, and it is worth stating plainly rather than leaving it to be discovered: the optional network streaming feature. It is switched off by default, it does nothing until you turn it on and enter the address of a peer, and when you do, the audio you route to it travels directly to that peer. Section 4 describes it in full. No other part of ToneSphere makes a network connection of its own.
2. What ToneSphere does with your audio
Audio entering ToneSphere is mixed, filtered and routed inside the running process and delivered to the outputs you have patched. It exists in memory buffers for as long as it takes to play it — a matter of milliseconds — and is then overwritten.
ToneSphere contains no recording feature. It writes no audio files. Nothing in the application saves, uploads, transcribes, analyses or fingerprints the sound passing through it.
This applies equally to the capture features:
- Per-application capture on Windows, which captures the audio of one process that you select by name or process id.
- System-wide loopback capture, which captures what is currently playing through an output device.
- Audio session detection, which lists the applications that are currently playing sound, so that you can pick one. This reads the operating system’s own audio session list; it does not read the contents of those applications or anything else about them.
Captured audio is treated exactly like any other input: routed, mixed, and discarded. It stays on your computer unless you have configured the optional network stream described in section 4 and routed it there yourself.
3. Information stored on your computer
ToneSphere stores its own settings on your machine, in ordinary files that you can read, edit and delete. Nothing in these files is sent anywhere.
| What | Where | Contains |
|---|---|---|
| Settings | settings.db in your user data folder (on Windows %LOCALAPPDATA%\Neural Nexus Studios\ToneSphere); defaults are read from config/default_config.yaml, which ships with the application and is never written |
Sample rate, buffer size, backend choice, port numbers, logging switches, interface language, the folders you asked to be scanned for plugins |
| Session | session.yaml in the same user data folder, saved as you work and when you close |
The same as a preset: your setup as you left it, so it comes back when you open ToneSphere |
| Presets | the presets folder in the same user data folder, or wherever you choose to save one |
Patch and mixer state as plain YAML, keyed on audio device names and identifiers; for each plugin in a chain, its file path, name, vendor, version and the plugin’s own saved state |
| Plugin scan cache | plugin_cache.json in the same user data folder |
For each VST3 module found in the folders scanned: its file path, size and date, and the names, vendors, versions and categories it declares |
| Log files | logs/tonesphere.log in the same user data folder, on by default (the enable_file_logging setting turns it off); it rotates, so it stays small |
Diagnostic lines about streams, devices, plugins and errors |
| Crash reports | logs/crash-*.log (named by date and time) in the same folder, written only when ToneSphere meets an error it did not expect |
The error message and the chain of calls that led to it, with the file paths of the code involved; the ToneSphere, Python and operating-system versions |
Some of these record names rather than only numbers, and it is fair to say so. Presets record the names of your audio devices — for example “Focusrite Scarlett 2i2” or “Speakers (Realtek Audio)” — and Windows’ identifiers for them, because a preset keyed on a device index breaks the moment you plug something in. Presets and the scan cache record which plugins you have and where they are installed, and a preset stores each plugin’s own saved state, whose contents the plugin decides. Logs may record device names, the file paths of plugins you loaded, and error messages; a crash report may too. All of them are local files under your control, and you decide whether to share one — for example by attaching it to an issue. Deleting those folders removes them, and uninstalling the application leaves nothing about you behind on any service, because there is no service.
4. The one feature that sends audio off your machine
ToneSphere can stream audio over a network to a peer that you nominate. This is off by
default (network.enabled: false in the configuration) and stays inert until you enable it
and supply an address.
When you do enable it:
- You choose the destination. The audio goes to the host and port you enter, and nowhere else.
- It is a direct connection between your machine and that peer. Neural Nexus Studios operates no server for ToneSphere. There is no relay, no intermediary, no cloud component, and no copy of your audio held by anyone. We cannot see this traffic, and there is nowhere for us to see it from.
- Only the audio you route there is sent. The network destination is an ordinary destination in the routing matrix. Whatever you patch into it is sent; nothing else is.
- The transport applies no encryption and no authentication. Audio travels as PCM in UDP datagrams, or over the TCP receive path, without a cipher and without a check on who is at the other end. Anyone able to observe the network path between the two machines could listen to that audio, and anyone able to reach your listening port could send audio into it. Use this feature on networks you trust, and do not treat it as a private channel across the public internet.
- The peer learns your IP address, as it must for any direct connection, and you learn the peer’s. That is a property of connecting two machines, not something ToneSphere adds.
The local REST API and WebSocket interface deserve the same candour. When you run
ToneSphere in server mode it listens on 127.0.0.1 by default, which means only your own
machine can reach it. If you change that host to 0.0.0.0 in order to control ToneSphere
from another computer, then any machine able to reach that port can also change your
routing and read your engine statistics; the interface has no authentication. That is your
decision to make, and the default is the private one.
5. Third-party plugins you load
On Windows, ToneSphere hosts VST3 effect plugins that you already own, on the device inputs and outputs you choose. Those plugins are software written by other companies, running inside the ToneSphere process at your instruction. Scanning for plugins loads each one briefly, in a separate process, to read what it declares.
A plugin can do things ToneSphere does not: check a licence over the internet, contact its vendor’s servers, write files, or collect information about your machine. What a plugin does is governed by that vendor’s own privacy policy and licence terms, not by this one. Neural Nexus Studios does not supply these plugins, cannot inspect their behaviour, and receives nothing from them. If you want to know what a plugin transmits, its vendor is the only authority on that.
6. Distribution channels
ToneSphere itself has no analytics, but obtaining a copy of it involves someone else’s service, and those services keep their own records:
- Microsoft Store. ToneSphere is being prepared for distribution through the Microsoft Store for Windows. Microsoft’s own data collection when you browse, acquire, install or update an application from the Store is described in the Microsoft Privacy Statement and is Microsoft’s, not ours. Microsoft provides publishers with aggregated reports — acquisition counts, ratings, reliability summaries — that do not identify individual users. That aggregate is the entirety of what Neural Nexus Studios receives about installations, and it is not personal information.
- GitHub. Releases are also published on GitHub. Downloading a file from GitHub, or opening an issue there, is processed under the GitHub Privacy Statement. Publicly filed issues are visible to everyone, including the GitHub username you write them under. Neural Nexus Studios sees the public issue and aggregate download counts, nothing more.
If a version of ToneSphere is ever sold, the sale is processed by the store or platform it is bought through, under that platform’s privacy statement. Neural Nexus Studios receives no card or bank details, and would receive only what that platform passes to publishers about a sale; no account with us is created or needed.
Neither channel reports anything to ToneSphere, and the application does not contact either of them while running. It performs no update check and makes no outbound request of its own.
7. Children
ToneSphere collects no information from anyone, of any age, so it holds no information about children. It is a general-purpose audio utility and is not directed at children.
8. Your rights over your information
Data-protection rights — access, correction, export, erasure, objection, withdrawal of consent — are exercised against an organisation that holds your personal data. Neural Nexus Studios holds none, because ToneSphere collects none. There is no account to close, no profile to export, and no record to delete.
The only ToneSphere-related data that exists is the local files listed in section 3, on your own computer. You can read them in a text editor, change them, or delete them at any time without asking anyone.
9. Security
Because nothing is collected, there is no central store of user data to breach. What remains is local, and worth understanding:
- Configuration, presets and logs are plain text and are not encrypted. Anyone with access to your user account on that computer can read them.
- The optional network stream is unencrypted and unauthenticated, as described in section 4.
- The local REST interface is unauthenticated and is bound to
127.0.0.1by default.
We would rather document those three facts than describe the application as secure and leave you to discover them.
10. Changes to this policy
This policy carries a version number and an effective date at the top of the document. A material change raises both, and the updated policy is published at the same address, so that the URL registered with the Microsoft Store always resolves to the current text. Earlier versions remain in the project repository’s history.
11. Contact
Questions about this policy, or about anything ToneSphere does with audio on your machine, should be raised as an issue in the project repository:
https://github.com/AvishakeAdhikary/tone-sphere/issues
That is the project’s contact channel. Issues are public, so do not include anything confidential in one.
12. Related documents
- Terms and Conditions — the agreement governing your use of the application itself.
- Terms of Service — what the application provides, how it is distributed, updated and supported, and the terms attaching to the optional network feature.