Even as someone really AI-forwards, there are just not enough selling points for me here. I almost never want to talk to an AI. I just don’t believe I’ll have a useful voice interaction. Maybe agents are here to fix that, but theres 30 years of really negative precedent from robot telephone bots to overcome, and I don’t think some new API is going to change that overnight
I almost never want to talk to an AI, because text is better for almost anything. But it's nice to have that "almost" corner case covered, no?
Not to mention all the current "telephone bots" applications that could benefit from something that has actual reliable STT and can accurately grasp a number you tell it first try, or hear a natural language description of what you want and immediately bypass listing the entire menu of options one by one.
I can’t tell if you are being serious or sarcastic. Assuming you are being real, and as someone who has to listen to boring youtube videos to help falling asleep, i have a few questions: how long is the story? What happens after the story ends? Like, does it turn off the voice feature or does it stay on and randomly speaks during the night?
hah I just tried it a few times. You have to choose a voice you like (Spruce for me) and then tell it continue a scene or chapter from one of your favorite books
Like for me it's the part in The Wind in the Willows where they're on the cart, before Toad becomes a car fanboy. I asked ChatGPT to stretch that part and make them stay on the road for longer, and it kinda did an OK job tbh
Yesterday I helped a maaaaybe 40 year old woman connect to the wifi at my local library so she could get on a zoom call. Made me realize most of the world is exactly like that.
Man, it's so important to read sometimes. I really hope HN people passing by this comment take a second to contemplate what's being said here.
The claim: "the models are tuned to align with one side of the issue, he is an article from ArsTech about it"
The reality: grok got mass-reported on X by pro-Israel accounts, leading to an automated suspension, which was undone by the X team shortly thereafter.
Absolutely nothing to do with the models aligning to one side or the other on this conflict. The claim is unfounded.
It's not that the models disagreed, the models reflected exactly the reality of the situation.
Sample output: "However, the definitive, legally binding determination of Israel’s responsibility under the Genocide Convention has not yet been made by the International Court of Justice"
The United Nations never made a ruling. You can have opinions about whether it should have, but the models are not lying to you, and at least the two examples above make very good efforts to explain the current accusations and claims made by both sides.
Stt/tts systems always seem to me so promising, but I pretty much never use voice to interface with a computer. Sometimes instead of typing on my phone, I use a voice dictation. I would be keen to use voice to control Claude code, but I've always felt that the way I speak is different from the way I write good prompts.
Fishing for anecdotes here, does anyone have any good tts/stt experiences?
I've always dreamed of having the ability to just talk to my computer (in the right circumstances) so I actually worked in the field for many years. The main reason I never use speech recognition today is because I have zero interest of sending recordings of my voice to the servers of some global corporations.
Running speech recognition and TTS locally is quite feasible, as projects like this one show.
If you want a local and open source option (MacOS only at the moment though), I've been happily using Keyscribe for dictation, which is built on Moonshine I believe.
I do a ton of coding (codex) with a tts/stt wrapper. During walks, cycling, in the car. Not every task is suited to this style of interaction, but many are. Long form codex replies are condensed, code blocks are suppressed all in the name of making it work for tts feedback. So it works best on well defined projects with guardrails, where you know the agent can perform well.
Talking on the phone is extremely dangerous? What are your thoughts on talking with a passenger? Is that something that people ought to stop doing? Drive in silence. 100% focus?
All the studies show it's not an over-reaction, you're 400% more likely to have an accident. It is extremely dangerous and it's not an over-reaction, the more people do this, the more people die. It's simple maths. If you're doing extended programming sessions, you're not paying attention to the road like you should be.
And no, talking to a passenger is not as dangerous as it's a different cognitive load.
It'll take you 20 seconds to google this, please do and stop putting everyone else in jeopardy.
In what way is talking on the phone a different cognitive load that talking to a passenger?
After googling the 8th or 10th result had a link to the study I assume you are referencing.
I just don't agree with you. I believe that there is risk when driving inherently although relatively low and a 400% increase from 0.0001 to 0.0004 does not really concern me.
I have been in one accident in 15 years and I was not at fault and it was unavoidable.
Many people engage in all sorts of distracted driving, texting, eating, driving while tired etc. which I would wager are more detrimental than speaking on the phone.
Having your one bone to pick being speaking on the phone just seems absurd and an odd message to preach when nearly everyone does it.
The difference between talking to a passenger versus using a hands-free phone is an interesting one. While they intuitively feel similar, it has been shown that speaking on the phone results in significantly worse performance. This is because a passenger shares awareness of the driving situation and surroundings. Their conversation naturally adapts to traffic conditions, often pausing or shifting topics to focus on the road, which can help with awareness.
Texting and eating are far more dangerous as they divert eyes from the road and introduce motor interference. Fatigue also severely degrades both attention and cognitive function.
Coding while driving does seem unnecessarily risky though. Unlike other non-motor distractions, it demands more cognitive engagement without necessarily moving your hands or eyes. It is far more demanding than a standard phone conversation, which has already been shown to impair driving.
You will make errors that you won’t remember because your brain is occupied with the code rather than monitoring the environment.
Your ability to recognize objects and react slows down. If a child runs into the road, that split-second delay could mean the difference between stopping or killing them. The same applies to sudden obstacles like kangaroos in my case. The delay in reaction could turn a near-miss into a fatal crash.
Most of the risk lies in the complex and unexpected situations. It's really not worth it, just listen to some music or an audiobook and enjoy your ride. Do coding later.
If you have any other psychology or cognitive-related questions to it I can try and answer. I'm not an expert in this specific area but we did some work on it at university.
That's fantastic. I have long, winding trails near me also and one of these days I also want to start prompting a coding agent on my bike with a headset. Do you recommend any particular type of headset?
Edit: never mind, I see you already suggested the Shokz OpenComm2 in another comment. Thanks!
I have my phone in my pocket, no screen interaction is required. I use a headset (Shokz OpenComm2) with wind muff (when cycling).
I made an Android app that listens for codex turn-complete or intermediate updates and plays them back to me. My answer is transcribed and pasted back to the relevant codex (tmux) session on the server (which I can select by voice) a tiny layer helps with things like /new, /plan, answer selection, etc.
I'm still wondering if this is something the big AI companies will come out with tomorrow, if it has business potential (it's extremely niche I think) or if I should put it on GitHub and build a community around it. This is what the Android app looks like:
https://imgur.com/a/qy8B2gthttps://imgur.com/a/C0eTPyn
I can't honestly think of a case where this would be remotely useful. This goes somehow beyond vibe coding to vibe interaction, where the only feedback comes via the AI. I'd love to see a concrete example of this working practice.
I've been building out the Android app that works as the wrapper between codex and me this way.
ADB debugging access over WireGuard. I ask to add a local wake phrase detection for the phrase "Hey codex" to start dictation. Codex tells me it thinks Vosk or OpenWakeWord might be a good fit, I voice select Vosk.
Codex starts working. Codex deploys (updates) the app on my phone and restarts it and I hear the task complete response.
I say "Hey Codex" to test and ask for some tuning/improvements, or move onto the next task. All while keeping my phone in my pocket.
But there are countless projects I work on this way. Eg, I got an email from person X, it says it encountered a bug when doing x and y. Codex reads the email (using nvidia/gliner-PII to hide PII data) and investigates the reason for the bug and proposes a solution. I ask it to implement the fix on my dev server and increase test coverage. I enjoy my walk and after say 7 minutes get a an overview of the fix and can decide to deploy to production.
I use the Hex[1] app on MacOs for near instant transcription with Parakeet V3. This is how I speak to Code agents at least 80% of the time. The idea is - I tend to be lazy if I need to type lots of details, so speaking lets me get into details that I otherwise wouldn’t and this helps give more useful context to the agent. This often tends to be an unstructured brain dump so I sometimes ask the agent to repeat back what it understood, so I can make sure, and this also likely helps it stay on track.
I used to use Handy but as of a few months ago it had stuttering and other issues so I switched to Hex. Even for relatively long minute long dictation, transcription is near instant and accuracy is more than sufficient, especially when talking to AI since it that can “read between the lines “.
Using voice with code agents is a huge unlock and I’m surprised to see some people I recommend it to, still resist it.
As for TTS I found it fun to make a voice plugin for Claude Code that uses PocketTTS, so it can give brief updates whenever it finishes a turn:
I (well CC and I) wrote a tts/stt pipeline for the CLI of CC. It's a lot more, immersive I guess, when I open my dev environment and it gives me a verbal walk through of what's going on.
(this inspired some more demo-y stuff I have where claude can manipulate the mouse and audit things it's built visually in conjunction with that). I'm sure this has already been wrapped up into some MCP framework, but it was fun to build it super early on and it just sort of works for me.
I don't use this in my day job, but it does feel very futuristic when I pull up my home lab.
I'm founder of ottex.ai, I use stt pretty much all the time when work with AI and quite often for communications to draft emails and chat messages.
I started ottex half a year ago after I tested gemini 2.5 flash native audio support. I was blown away by the quality of transcripts and decided to built an app to use it myself.
Currently the default model in the app is Gemini 3 flash, but you can connect to 9 providers and God knows how many models to play with.
I would suggest you to try this models for ai prompting:
Industry leading Interactive Voice Response systems have become very good at filling in ambiguous information from context, and modulating pronunciation to Ape emotional information.
However, being able to interact with these natural language systems in uncontrolled settings is still a fools errand. For STT, there is also regional dialect, slang, and individual differences.
Witnessing blind users hit unrecognizable reading-speeds on old Gordon 8 TTS systems was surprising. I learned people adapt to imperfect systems pretty quickly. =3
What I want is a 1940s style “taking dictation” where the words I say go through a step where the goal is to create the text Im imagining. So if I say “… very significant, actually just significant…” what Claude Code receives is “…significant…”.
I built this myself with whisper -> “secretary” prompt -> Claude Code, but having the first two steps be interactive is really what I would want.
I use dictation to drive Claude code frequently, and it’s never had a problem with stream of consciousness and retroactive correction. Maybe try just direct voice and see if you notice any difference versus pre-cleaning?
Claude Code's speech recognition works so well for me, I was blown away the first time I tried it. I wish I knew what model they were using (I assume it's not in house since they've never talked about it).
I acknowledge this may just mean I haven't tried enough modern voice recognition systems. But I've used Whisper and I don't think it works nearly as well for real-time speech.
(I still don't tend to use voice mode in Claude Code because I find typing more comfortable.)
This reply helped me think more about what I want. I do my thinking when crafting the prompt, and I want to be able to do several passes. So the first draft I like to dictate and be able to live edit with my voice, but then I want to do a subsequent pass at the keyboard after the full prompt is written.
https://handy.computer is the goto. It can even add an LLM pass (optionally) so if you say "remind me to buy three, actually, four, eggs" it'll type "remind me to buy four eggs".
For STT, wispr flow has a generous free tier. For TTS, I have Claude read out loud what it just finished as a stop hook, so I know which claude finished up.
One of my side projects is a tool that lets you control your entire system with STT. It's built on Whisper and supports hot swapping custom profiles, so you can add easy commands for any software.
I intend to use it to work on low stakes vibe coding projects while I'm doing other stuff. Todays LLMs are a lot better at interpreting rambling dictation with mid-message corrections.
There are a few paid programs out there that do the same, but they made my vibe slop sense tingle and are not aimed at development.
If the corporate berries are really so bad, the invisible hand will push the company in the direction of society's aggregate wallet vote. Sounds like most people are fine with them. Outside of truly autocratic systems, sounds like these berries are WAI.
* The people are not fine with bad strawberries but have no other choices available
* The people are not fine with bad strawberries but can't afford better choices
* The people are not fine with bad strawberries but they don't know good strawberries
* The people are not fine with bad strawberries but they're cheap enough to ship and sell that there's no economic case for good strawberries, so no one close enough to buy from will sell good strawberries to them
"The market seems fine with it" is kind of a lazy thought terminating cliche answer. What if the invisible hand of the market is pushing strawberry producers towards the outcome "society no longer values this enough to buy it" in which case the aggregate wallet vote will be zero?
The people are not fine with bad strawberries but they don't know good strawberries
You most definitely get this phenomenon with tomatoes. There’s little demand for actually good tomatoes, because most people don’t even know what a good tomato tastes like at this point.
This applies to countless things, but tomatoes are a prime example because they deteriorate so quickly once picked relative to other fruits I guess. So they have completely bred the flavor out of them in a quest to achieve something that looks good on a supermarket shelf.
This is a phenomenal example I hadn't even considered, because I have been affected by this kind of "invisible hand of the market" negative quality spiral.
The older generation here remember good tomatoes, so they continue to buy bad tomatoes but will complain every time they eat them about the quality. I get told a lot about heirloom varieties and how good they are in comparison.
I grew up with modern tomatoes. I've never tried an heirloom so I can't compare, but I don't recall ever eating a good tomato, so I just don't buy them. The market has moved itself into a position that shrinks its own demographic.
I see people constantly make this argument, and honestly I think it’s BS. I grew up eating tomatoes from my grandparents garden, and I’ve lived and traveled all over the world. I’ve grown tomatoes, bought them from roadside farmer stands, bought them at grocery stores, and had them in everything from hole in the wall restaurants in developing countries to Michelin three star restaurants on multiple different continents.
Today’s grocery tomatoes are fine. And my grocery stores generally have 5-10 varieties too.
Yes, you can get better ones, but not to where it’s some religious experience that will forever ruin grocery store tomatoes.
On top of that, most people really don’t care that much, not because they don’t know any better, but because the cost and convenience factor trumps the slight subjective increase in quality. I doubt most people could even tell the difference between two tomatoes of the same type and ripeness if one came from the grocery store and the other from a backyard garden.
I’ll grant that most non-local tomatoes have always been bad by definition, because they’re picked while green so they don’t rot before reaching the store.
Plum tomatoes absolutely did not used to be this bad, though. They are SO mealy now. Horrible. Beefsteaks are mealier as well. Those Campari tomatoes are pretty good year-round, though, I have to admit.
This is all in the NE USA, FWIW. I don’t know the tomato situation elsewhere.
I doubt most people could even tell the difference
between two tomatoes of the same type and ripeness
if one came from the grocery store and the other from
a backyard garden.
Yeah, and I would run as fast as Usain Bolt if we woke up with the same body one day.
But that kind of the thing. They would almost never be the same ripeness because outside of local tomato season the tomatoes are picked while unripe, and then they “shelf-ripen” in transit because ethylene gas etc. That’s always been an issue, of course, and hasn’t changed over time.
The other issue is breeding - the continual breeding for appearance rather than flavor. Maybe we’re all imagining that one.
Isn't the point that we don't grow the good varieties any longer because they don't survive freight? It doesn't matter if you bought the tomato seeds from Harrods and grew them in your lush orchard if they're the same lineage bred for shipping hardiness over all else.
1. Store-bought tomatoes are nearly always bred for shippability and appearance over flavor
2. Store-bought tomatoes are picked when unripe, so they shelf-ripen during transit and at the store, which is highly inferior to ripening on the vine for flavor
For the first issue... you can buy heirloom tomato seeds at any major hardware or garden store in America.
For the second issue... even the typical tomato breeds will taste great if you grow them yourself and let them vine-ripen till they're ready to eat.
This reminds me of the Jewish Sabbath. Here in Tel Aviv, Saturday means no shops, much fewer restaurants, less programming on TV, gyms/etc open late/close early.
The parks and beaches are full of people just existing.
On Saturdays, in the town of Surfside, I would frequently see many Orthodox Jewish people at the beaches and cheerfully going for walks and so on. A pleasant and wonderful atmosphere.
Reminds me of my friends in Gaza, where every day means no shops, no restaurants, no home, no medicine, no food, no water, no hope. The beaches are full of people just trying to exist.
The person you are talking to did not personally commit genocide in Gaza. Just as I, an American citizen, did not declare war on Iran.
In fact, I think the war in Iran is a stupid and immoral thing to do. It's possible that the person you are responding to feels the same about the Israeli government's genocidal actions in Gaza.
However, you did not bother to find out. When you judge someone before knowing them, it is called "prejudice". Pre-judging.
Did the comment you replied to actually accuse them of any of those things? I don't see them doing that. They juxtaposed the two situations, but they made no accusations nor casted any blame. There isn't even any prejudice I can see. Was the comment edited or something? I don't see how your comment makes sense as a reply to its parent.
> I don't see how your comment makes sense as a reply to its parent.
But you can see why the Gaza response made sense? It was completely irrelevant to this discussion. I’m not sure how you’re missing the heavy sub text here.
Friendly reminder that, for all that I despise Israeli politics, "existing in Tel-Aviv" isn't a crime or an aggression against Palestinians.
We can think ill of the Israeli state without jumping to "fuck you for living in Israel and having nice things" as soon as someone mentions their city name.
The UN classified it a genocide. Is the latest to call that a marketing ploy or Russian disinformation and say it's not OK or hip or fashionable to question our allies? Evil productized?
I did something similar. My computer crawls lite.cnn.com each Saturday. I feed it all into Gemini who composes a "front page" with links in HTML. Then the whole thing is converted to a PDF and uploaded to my Google drive with the day's date as the title. My Boox reader (some Chinese company) is synced to my Google Drive and I just open it from there. I didn't even code any of this, Gemini did.
It's a nice thing to read on Saturday morning with a coffee.
