Speech to Text on iPhone vs. Pixel • Hacker News Board

Re: Speech to Text on iPhone vs. Pixel

#31

Post by walterbell » Wed, May 27, 2020, 4:10 PM UTC

Some hackers have been trying to reuse Google's offline speech recognition models within other software toolkits, https://hackaday.io/project/164399-android-offline-speech-re...

> Especially the offline part is very appealing to me, as it should to any privacy conscious mind. Unfortunately this speech recognizer is only available to Pixel owners at this time. Since GBoard uses TensorFlow Lite, and the blog post is also mentioning the use of this library, I was wondering if I could get my hands on the model, and import it in my own projects, maybe even using LWTNN.

Recent (May 2020) news suggests that these models may be coming to Chromium, which would make them widely accessible for offline transcription and dictation, e.g. WebRTC or video captions, https://hackaday.io/project/164399-android-offline-speech-re...

> Google is building speech recognition into Chromium, to bring a feature called Live Caption to the browser. To transcribe videos playing in the browser a new API is slowly being introduced: SODA ... What is especially interesting is that it seems it will be using the same language packs and RNNT models as the Recorder and GBoard apks

In Aug 2019, the Live Transcribe engine was open-sourced, https://github.com/google/live-transcribe-speech-engine & https://opensource.googleblog.com/2019/08/bringing-live-tran...

Re: Speech to Text on iPhone vs. Pixel

#32

Post by caycep » Wed, May 27, 2020, 4:16 PM UTC

granted neither google or apple really works well with my voice for some reason...

Re: Speech to Text on iPhone vs. Pixel

#33

Post by sangnoir » Wed, May 27, 2020, 4:17 PM UTC

ksec wrote: post #14

yalogin wrote: post #11

Wow does anyone know why it’s slow on the iPhone?

iPhone is doing this online, sending data packet back to Apple. Pixel is doing this completely offline.

I'll add another non-technical reason: ML is a Google core-competency, and has been for some time now[1], but it is/was not central to Apple.

1. Google 411 was introduced 13 years ago, in 2007, and look at the ML papers published and techniques Google pioneered since then.

Re: Speech to Text on iPhone vs. Pixel

#34

Post by whereistimbo » Wed, May 27, 2020, 4:51 PM UTC

WhyNotHugo wrote: post #8

I'm curious about the privacy policies behind both engines and their development. Google's gotten an advantage in many fields by simply trampling over user's rights. Apple has been more respectful of user rights, and that's kind of made life harder for them too.

Not necessarily user data. Google own reCAPTCHA and YouTube. Think about that.

Re: Speech to Text on iPhone vs. Pixel

#35

Post by gonational » Wed, May 27, 2020, 4:51 PM UTC

I will never support google in anyway, at least in tension fully, but Apple voice to text is absolutely a palling. I just dictated this message here is my iPhone, so I will leave it as is without fixing the fuck ups.

Re: Speech to Text on iPhone vs. Pixel

#36

Post by tialaramex » Wed, May 27, 2020, 5:02 PM UTC

Similar offline versus online privacy thing for music recognition.

My Pixel is sat here charging. Billie Eilish is singing in the background, by the time she sings "I should have known..." it has recognised the music and passively displayed "No Time To Die by Billie Eilish" on the lock screen.

Google's team were building an ML model for music recognition and they realised oh - the smallest model fits on a phone. We don't need to spin this up as a cloud service we can just deliver it to the phones of people who want it as a feature. "Now your phone can tell you what that music is".

Of course down the road this awareness of context allows even cleverer agent interaction. "Which Bond was that?" you ask as Adele sings "Skyfall". "In the movie Skyfall, James Bond was played by Daniel Craig".

Re: Speech to Text on iPhone vs. Pixel

#37

Post by causality0 » Wed, May 27, 2020, 5:10 PM UTC

For me it seems like Google voice typing is getting worse and worse. It capitalizes random Words in a sentence Like this. It specifically ignores the word "o'clock".

Re: Speech to Text on iPhone vs. Pixel

#38

Post by paulgb » Wed, May 27, 2020, 5:10 PM UTC

cl0rkster wrote: post #4

There are already some novel applications of this. It is extremely useful to use Google's screening service for unknown numbers that might be a call I'm expecting from a new number. I can't recall the last time I actually listened to a voicemail. I know that at least voicemail transcription works elsewhere, and I can't speak to how well, but I do know that on the Pixel, transcription is really quite impressively accu…

Call screening has saved me so much time that it's one of the main things stopping me from switching to iPhone.

Re: Speech to Text on iPhone vs. Pixel

#39

Post by oh_hello » Wed, May 27, 2020, 5:22 PM UTC

An an iPhone user, I very often find speech recognition far too slow. It is to the point I rarely use the feature because it often causes frustration. The Google performance here is amazing.

Is this a feature only of Pixel phones?

Re: Speech to Text on iPhone vs. Pixel

#40

Post by tinus_hn » Wed, May 27, 2020, 5:28 PM UTC

tialaramex wrote: post #36

Similar offline versus online privacy thing for music recognition. My Pixel is sat here charging. Billie Eilish is singing in the background, by the time she sings "I should have known..." it has recognised the music and passively displayed "No Time To Die by Billie Eilish" on the lock screen. Google's team were building an ML model for music recognition and they realised oh - the smallest model fits on a phone. We d…

More likely the clever part will be you will get more ads for spy cameras and the like.