Avatar
Max
b7ed68b062de6b4a12e51fd5285c1e1e0ed0e5128cda93ab11b4150b55ed32fc
Praxeologist ~ Cryptoanarchist ~ Cypherpunk nostr:naddr1qqgrvvpjx33kgvtrxgen2vfcvcurjqghwaehxw309aex2mrp0yhxz7n6v9kk7tnwv46z7q3qklkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qxpqqqp65wvem775
Replying to Avatar Merlin

nostr:nprofile1qy88wumn8ghj7mn0wvhxcmmv9uq3wamnwvaz7tmjv4kxz7fwv9a85ctdduhxuet59uqzpdlddzcx9hntfgfw28749pwpu8sw6rj39rx6jw43rdq4pd276vhu7r8p0e is there a place to report bugs or suggest improvements for White Noise besides the Signal chat? It would be great to see what’s already on your radar.

I’ve been using the app for a while, have noticed a few things, and would like to contribute.

Here on nostr is fine, also GitHub issues for less clutter, and the White Noise support account in the app itself.

Thanks!

Replying to Avatar Final

We're developing our own implementations of text-to-speech and speech-to-text to use in #GrapheneOS which are entirely open source and avoid using so-called 'open' models without the training data available. Instead, we're making a truly open source implementation of both where all of the data used for it is open source. If you don't want to use our app for local text-to-speech and speech-to-text then you don't need to use it. Many people need this and want a better option.

We are working on TTS first then SST. The TTS training data is LJ Speech https://keithito.com/LJ-Speech-Dataset/ and the model used is our own fork of Matcha-TTS.

If people want they can fork it and add/remove/change the training data in any way they see fit. It's nothing like the so-called "open" models from OpenAI, Facebook, etc. where the only thing that's open are the neural network weights after training with no way to know what they used to train it and no way to reproduce that.

Many blind users asked us to include one of the existing open source TTS apps so they could use it to obtain a better app. None of the available open source apps meets our requirements for reasonable licensing, privacy, security or functionality. Therefore, we've developed our own text-to-speech which will be shipping soon, likely in January. We'll also be providing our own speech-to-text. We're using neural networks for both which we're making ourselves.

Would this include only the TTS engine, or also an app to read texts and show them in the UI?

iOS doesnt allow client side notifications, so we first have to invent a private notification server scheme, that's MIP 5 https://github.com/marmot-protocol/marmot/pull/18

On Android the client keeps running in the background and checks for new messages to notify you every 30 seconds.

Replying to Avatar Final

We're developing our own implementations of text-to-speech and speech-to-text to use in #GrapheneOS which are entirely open source and avoid using so-called 'open' models without the training data available. Instead, we're making a truly open source implementation of both where all of the data used for it is open source. If you don't want to use our app for local text-to-speech and speech-to-text then you don't need to use it. Many people need this and want a better option.

We are working on TTS first then SST. The TTS training data is LJ Speech https://keithito.com/LJ-Speech-Dataset/ and the model used is our own fork of Matcha-TTS.

If people want they can fork it and add/remove/change the training data in any way they see fit. It's nothing like the so-called "open" models from OpenAI, Facebook, etc. where the only thing that's open are the neural network weights after training with no way to know what they used to train it and no way to reproduce that.

Many blind users asked us to include one of the existing open source TTS apps so they could use it to obtain a better app. None of the available open source apps meets our requirements for reasonable licensing, privacy, security or functionality. Therefore, we've developed our own text-to-speech which will be shipping soon, likely in January. We'll also be providing our own speech-to-text. We're using neural networks for both which we're making ourselves.

Amazing to hear, thanks!

Private Information Retrieval can't solve Nostr's query privacy problem. The compound predicates, range queries, and federated trust model break it. But there's another approach: run the relay inside a secure enclave where even the operator can't see your queries. This post examines what TEEs actually offer, what they don't, and whether the trust trade is worth making. nostr:naddr1qqgr2cfcvcmrjwfcvdjnse35xuckvq3qklkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qxpqqqp65wrtjqdc

The state claims to represent you, protect you, serve you. It does none of these things. It is simply the organization of predation, dressed in the language of public service. This post dissects the state: what it is not, what it is, how it survives, how it grows, and what it fears. nostr:naddr1qqgrsdnpxumkgwfevscnyetxxccrgq3qklkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qxpqqqp65wx0zqej

Could Private Information Retrieval let Nostr clients query relays without revealing who they follow? In theory, yes. In practice, Nostr's filter syntax raises hard cryptographic questions that don't have answers yet. This post asks those questions. nostr:naddr1qqgr2dnrxvcx2ctpxu6nxenpxeskyq3qklkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qxpqqqp65wu273pc

If you want minimal price, go with the 8a, for most performance the 9 pro or 10 pro.

The 9a comes with better build quality and one more year of updates (always 7 years after launch)

Here two articles to help set it up

nostr://nevent1qqsv93qudtfgpcnrwvt6xfrj6y752cevftankd0k6lmv8hdgs94e7zcpzamhxue69uhhyetvv9ujuct60fsk6mewdejhgtczyzm7669svt0xkjsju50a22zurc0qa589z2xd4yatzx6p2z64a5e0cqcyqqqqqqg4xl25t

nostr://nevent1qqsg5vs07t0y7q5jw3tak264mt5f984enp4dkh2d5ygwx2dsx7sjzfgpzamhxue69uhhyetvv9ujuct60fsk6mewdejhgtczyzm7669svt0xkjsju50a22zurc0qa589z2xd4yatzx6p2z64a5e0cqcyqqqqqqg0q7qk3

The West is not "stagnating." It is being destroyed.

Decades of taxation, inflation, and regulatory plunder have consumed the capital that took generations to accumulate. Europe now runs on fumes, its industrial base gutted, its currency debased, its people impoverished by design.

The path back to prosperity requires exactly one rule: **do not steal**. This principle, applied to governments as strictly as to criminals, would reverse the collapse almost overnight. The question is whether anyone has the courage to admit the obvious before there is nothing left to save. nostr:naddr1qqgxxdfjv9jxvdpsv3jnxve3xgcnzq3qklkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qxpqqqp65wf9ca8t

The Pixel 8's Tensor G3 chip pairs with an upgraded Titan M2 security coprocessor to isolate sensitive cryptographic operations, while introducing Memory Tagging Extension (MTE) support for runtime detection of memory corruption vulnerabilities and significantly hardening the cellular baseband firmware with bounds sanitizers, integer overflow protection, stack canaries, and control flow integrity to reduce what has historically been a major attack surface.

That's a major improvement compared to the 7 or earlier generations.

For two decades, we accepted a bargain: let Apple and Google verify every developer with government papers, and they will keep us safe from malware.

The bargain was always a lie.

Fake apps still steal millions while real developers get banned at the whim of bureaucrats in Cupertino and authoritarian censors in Moscow.

nostr:nprofile1qqs83nn04fezvsu89p8xg7axjwye2u67errat3dx2um725fs7qnrqlgpzamhxue69uhhxetpwf3kstnwdaejuar0v3shjtc3uszjr, a Nostr-based app store, offers a different model: one where developers sign their own releases, users verify through social trust, and no passport is required to publish code. This is not a feature request. It is a return to the original promise of the internet. And nobody can stop us building it.

nostr:naddr1qqgrsdfcx4nrjdpsvs6njenyxsunvqgwwaehxw309ahx7uewd3hkctczyzm7669svt0xkjsju50a22zurc0qa589z2xd4yatzx6p2z64a5e0cqcyqqq823ccq4deh