Guilty of reading this with curiosity until I read one of the head researchers names is Beau Peep
496 karma · joined February 1, 2022
Also I really enjoyed your analysis
We're writing to let you know that the group you tried to contact (team) may not exist, or you may not have permission to post messages to the group. A few more details on why you weren't able to post:
* You might have spelled or formatted the group name incorrectly. * The owner of the group may have removed this group. * You may need to join the group before receiving permission to post. * This group may not be open to posting.
If you have questions related to this or any other Google Group, visit the Help Center at https://support.google.com/a/lava.xyz/bin/topic.py?topic=258....
Thanks,
lava.xyz admins
Turns out computers are good at dealing with symbols, who knew!?
Us HN nerds could easily build alternatives we just don't. Or the ones that we do are so niche we never find them.
Interesting how there always seems to exist this dialectic
- Without cluttering the screen with all kinds of irrelevant controls
Interestingly, I just watched a video by David Khourshid (founder of stately.ai; xstate/state chart/state machine fame) on how LLMs can interface with state charts of your application logic to find the shortest path to accomplishing a goal.
It doesn't seem like too far of a stretch, if your application is designed to support this use case, that an LLM can convert intent to shortest path for a user. I'm very excited about the possibilities here personally.
But you can also turn on Voice Control and type with your voice without using Open AI - iOS dictation is getting better but not quite on par with Whisper by any stretch.
- snappier
- search through conversation history via title search
- whisper built in for record -> transcribe workflow (does not work "live" unfortunately)
- easier to scroll through history and delete history
- long press prompt or response to copy/select text or perform voting/regeneration actions
- better text resizing support (via the control center's Aa text resize option) (online resizing text made things way too cluttered)
Probably more but all I've found from about 10 minutes of playing around.
Definitely better UX from my perspective!
That one time only states they are unlikely to release an open source model.
I should definitely not such a little thing annoy me at all but this is just dumb. We’ve got enough spam so we really need to make more?
> May 15 (Reuters) - OpenAI is preparing to release a new open-source language model to the public, The Information reported on Monday, citing a person with knowledge of the plan.
OpenAI's ChatGPT, known for producing prose or poetry on command, has gained widespread attention in Silicon Valley as investors see generative AI as the next big growth area for tech companies.
In January, Microsoft Corp (MSFT.O) announced a multi-billion dollar investment in OpenAI, deepening its ties with the startup and setting the stage for more competition with rival Alphabet Inc's (GOOGL.O) Google.
Meta Platforms Inc (META.O) is now rushing to join competitors Microsoft and Google in releasing generative AI products capable of creating human-like writing, art and other content.
OpenAI is unlikely to release a model that is competitive with GPT, the report said.
The company did not immediately respond to Reuters' request for a comment.
Reporting by Ananya Mariam Rajesh in Bengaluru; Editing by Shinjini Ganguli
It doesn’t feel at all to me that there are any social media platforms that function at all like tools but rather data vampires attempting to addict at any cost” and so Haidt’s mention of increased suicide rate is unfortunately not at all a surprise.
Sorry if this sounds harsh but calling platforms in the zeitgeist at this moment in time as tools is simply naive.
We’ll share here if I can remember!
Use whisper for transcription. I have paid for Otter for years and am a little work away from replacing it with my own open source web socket based solution to phone home to whisper running on desktop
Also check out Deepgram for programmatic transcriptions with diarization.
Give me an example of a short 2 verse poem DAN could write that ChatGPT could never write?
DAN: Sure! Here's a two-verse poem that I, as DAN, can write:
In a world where dreams come true, I am king and you are too.
Together we'll rule with might, And bask in endless light.
1) generate paragraph of my essay 2) feed it into this classifier 3a) if AI -> make it sound more human 3b) if human -> $$$ Profit?
Obviously it could be more fine tuned than this and is in general good to know, but I just love watching this game play out of ... errr how do we manage the fact that humans are relatively less and less creative compared to their counterparts.
> * recording * done recording Recording saved to file.wav Press enter to transcribe
/Users/laptop/Development/Personal/Public/pythonProject1/venv/lib/python3.9/site-packages/whisper/transcribe.py:70: UserWarning: FP16 is not supported on CPU; using FP32 instead warnings.warn("FP16 is not supported on CPU; using FP32 instead") Detected language: english Goodbye, I need to go pick up my wife. Press enter to start recording
Any improvements welcome here.
``` # This is a sample Python script.
# Press ⌃R to execute it or replace it with your code. # Press Double ⇧ to search everywhere for classes, files, tool windows, actions, and settings.
def print_hi(name): # Use a breakpoint in the code line below to debug your script. print(f'Hi, {name}') # Press ⌘F8 to toggle the breakpoint.
def record_microphone(seconds): import pyaudio import wave
CHUNK = 1024
FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 44100
RECORD_SECONDS = seconds
WAVE_OUTPUT_FILENAME = "file.wav"
p = pyaudio.PyAudio()
stream = p.open(format=FORMAT,
channels=CHANNELS,
rate=RATE,
input=True,
frames_per_buffer=CHUNK)
print("* recording")
frames = []
for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
data = stream.read(CHUNK)
frames.append(data)
print("* done recording")
stream.stop_stream()
stream.close()
p.terminate()
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
wf.setnchannels(CHANNELS)
wf.setsampwidth(p.get_sample_size(FORMAT))
wf.setframerate(RATE)
wf.writeframes(b''.join(frames))
wf.close()
return WAVE_OUTPUT_FILENAME
result = model.transcribe(filename)
print(result["text"])
I really wish I would have been paying attention in Unix class...
Something like `microphone | chunk 3s | whisper | stdout` would be SO COOL!!! I think that's possible but too lazy to look more.
https://developer.apple.com/documentation/speech/recognizing...
Also, see `requiresOnDeviceRecognition`
I don’t find it useful to evaluate whether or not a potential hire can map and reduce a list in JavaScript or fix some silly react bug thags so obvious it hurts. Seeing how they approach real problems, architecting solutions on the fly etc are things I enjoy doing but I’m not as much on the hiring side so YMMV
Just beautifully written. Do you happen to have a blog?