Amazon Bedrock Is Now Generally Available

https://aws.amazon.com/blogs/aws/amazon-bedrock-is-now-generally-available-build-and-scale-generative-ai-applications-with-foundation-models/

by epberry • 3 years ago

33 10 3 years ago

aws.amazon.com

paulddraper• 3 years ago

Tangential, but I kinda miss the days when this would be SFM, Simple Foundation Models.

bguberfain• 3 years ago

It says it has support for Llama 2, but in a deep page you can read that it is "coming soon". Anyway, good to see support for serverless inference and painless train of Llama 2 models!

willtemperley2• 3 years ago

Interesting timing given CloudFlare just released Workers AI, which actually has Llama 2 [1], not just coming soon. Great to see some competition.

willtemperley2• 3 years ago

[1] https://blog.cloudflare.com/writing-poems-using-llama-2-on-w...

leoqa2• 3 years ago

Asking because I’m lazy: if I need to transcribe audio in real-time, is there a state of the art model I can plug into?

CrimsonCape• 3 years ago

https://github.com/ggerganov/whisper.cpp

https://github.com/Const-me/Whisper

I had fun with both of these. They will both do realtime transcription. Bit you will have to download the training data sets…

theossuary• 3 years ago

I saw whisper recommended, but I was curious how it compares to the other robust ASR systems (like nvidia's Nemo + Riva). I found this Twitter thread that seemed relevant: https://nitter.net/lunixbochs/status/1574848899897884672

Long story short, it depends on what you want to use it for. Different models and different training sets can help optimize for different things. Also, if you're in a domain with very uncommon speech patterns (think doctor shorthand or radio lingo), you'll need to understand how difficult it will be to customize generated models to do better in your space. I think Nemo + Riva does well at this; but I'm not as familiar with other options.

leoqa2• 3 years ago

Yeah I fine tuned a model 2 years ago but it was a big pain and performance didn’t get better than 85%

crakenzak• 3 years ago

whisper from OpenAI works great for me.

ranman• 3 years ago

If you want to use this but are tired of waiting for the SDKs to get updated you can do this:

```

curl -sS https://cdn.caylent.com/bedrock_new.zip > bedrock.zip

mkdir -p ~/.aws/models/

unzip bedrock.zip -d ~/.aws/models

```

Amazon Bedrock Is Now Generally Available

Related Stories