Hacker News Reader

Comment by n8cpdx | Hacker News Reader

Does anyone know of something like Otter but offline with tolerable performance?

Otter is the only thing approaching affordable for individuals, but I don’t really trust it given they’re playing fast and loose with user data.

View on HN

Pixel phones have a voice recorder that works offline and does this. I've used a bootleg apk I got off xdadevs of it on my samsung phone, and it works well. But for privacy or opsec that's probably even worse than a saas option.

Ostensibly almost all newer/flagship iphone/android devices can do offline speech to text with passable accuracy. On Android it is (I suspect intentionally) nerfed to an accessibility feature called "live transcribe." It transcribes everything star wars style on the screen, but you can't save the audio or the transcription, you have to copy/paste the output which just disappears when you close the app. It's arbitrary and infuriating that it's like this.

I'm generally a "hater" of the cloud, and I have to admit that Otter is very good. I used it speaking with someone using airpods on a shitty internet connection, with no noticeable drop in accuracy. Even spelled uncommon last names correctly.

What I don't get is why none of these free apps can take a pre-recorded audio file as the input. Even for the paid services, this seems to be more limited than using the functionality 'live.' Is there some technical reason for this, or is it a sales department decision?

> But for privacy or opsec that's probably even worse than a saas option.

Probably not, the attack vector of an apk of dubious provenance is probably that it sends all of your transcriptions to some 3rd party server. There's some non-zero probability that it does that, and it's possible you could run it in a network-less sandbox somehow to mitigate it.

But with the saas service, you know for a fact it's going to a 3rd party server, who can store it for as long as they want. And if they aren't hacked now, they could be in the future.

That analysis would be a lot more compelling if it didn't assume that all 3rd parties are the same.

A 3rd party sending around a sketch apk that calls back to them is _definitely_ going to try to do something bad with it.

Otter is unlikely to do something bad with it (it'd be bad for their business). So you should really compare the odds otter gets hacked vs the odds the apk is compromised.

Actually, I still have the bootleg apk installed, because sometimes you've just gotta live a little, right? Also happen to have the Otter app. Just checked netguard and guess which one has connected to the internet 6 times since I last opened it.

Silero[0] seems to have decent performance (although you will have to some minimal coding). I believe there are better ones if you're willing to tinker a bit more.

[0]: https://github.com/snakers4/silero-models

I recently found this new service called revoldiv.com few months back here on HN and I have been using it to transcribe and edit some of my videos. They are based in US and use AI to process their audio and transcripts. I believe the audio and transcripts are deleted once you close out the browser. It’s the only service that I found so far with that level of capability to transcribe/edit for free and seems safe… at least right now.

https://revoldiv.com/

I've had great success doing transcription on-prem with Speechmatics: https://www.speechmatics.com They offer both a VM and a Docker image that you can run behind your firewall, and even license offline if you really need to. I use it for generating closed captions of videos, but you could build a transcription tool out of it as well. Their engine's accuracy is the best I've found out there, too, which is a nice bonus.

The Recorder app on Pixel phones offers on-device transcription. I'm not sure if it's available on other phones though.

I haven't used it, but Coqui might be good:

https://coqui.ai/

You can use Amazon Transcribe. The output is just a block of text but that's fine for a lot of interviews you're just going to grab a few sentences from.

There are some various other services along similar lines to Otter.ai but don't remember names off the top of my head.

I don't think Amazon Transcribe works offline.

No but the parent's second line seemed to be a complaint about Otter specifically. I would probably trust Amazon to have good privacy practices generally though there's still some sensitivity line beyond which I wouldn't use an online service period even if they seem trustworthy in general.

> something like Otter but offline with tolerable performance?

Shorthand

Reply on news.ycombinator.com