NVIDIA just released a new open source transcription model, Nemotron Speech ASR, designed from the ground up for low-latency use cases like voice agents.
Here's a voice agent built with this new model. 24ms transcription finalization and total voice-to-voice inference time under 500ms.
This agent actually uses *three* NVIDIA open source models:
- Nemotron Speech ASR
- Nemotron 3 Nano 30GB in a 4-bit quant (released in December)
- A preview checkpoint of the upcoming Magpie text-to-speech model
These models are all truly open source: weights, training data, training code, and inference code. This is a big deal! Jensen said in the CES keynote yesterday that he expects open source models to catch up to proprietary models this year in a number of categories. NVIDIA is putting their weight behind making this happen. (As Alan Kay said, the best way to predict the future is to invent it.)
The code for this agent is open source too, of course. You can deploy it to production with @modal and @pipecat_ai cloud, or run locally on an @nvidia DGX Spark or RTX 5090.
@Lenskart_com@peyushbansal
I am new to wearing glasses and when I wear them for a long time, I feel uncomfortable. I take my glasses off but then they always fall from my hands.
Can we add holes in the glass frame so people like me can put a thread around (old school style)?
@pankaj_cic The architecture looks like a custom RAG application but fail to explain how it is going to recognise struggling student. IMO, a larger part of making the student feel better is to have empathic responses. Think about adding empathic awareness to the model.
Looking for 3 full-time Ph.D. students for funded projects.
Qualification: BTech/MTech completed; good math, coding skills in DL. Position will be filled when a suitable candidate is found; early start; Apply here: https://t.co/yJcblRZbhv
@SBILabIIITD#research#phdposition
#RecruitmentAlert Ph.D. Students, Project / Research Associate, Interns, Postdoc! Social Media Analysis, Computational Social Science, NLP, #DataScience 4 Social Good
Apply https://t.co/L2gzxAbdAF or Scan the QR code.
Please review https://t.co/oVf7LcDLGg before applying.
@ManvithReddy18 will be presenting our work "A Virus Has No Religion": Analyzing Islamophobia on Twitter During the COVID-19 Outbreak at @ACMHT on 1st September between 16:30-17:30 hrs (GMT +1)
Please drop by if you are attending HT'21 🥳
\c @shradhasgl@reallysaurabh@ponguru
Some friends and I have updated this. One can now access updated info on availability of COVID Hospitals ICU/Ventilator beds in Delhi and NCR.
🔗 https://t.co/F1k31PsgA8
@cheekusahoo They both are subjective. Only difference is that you get to choose if something is wrong or right, and whether it is illegal or legal is imposed upon you by the law. For example, you might think killing people is wrong and illegal, but it can be legal under certain circumstances
I wrote another blog on Data Leaks. I kept the language as simple as I could so that non-tech people can also understand the underlying issue. Share if you find it useful. 🙂
https://t.co/G3wnpUGNTX
In only 1.5 hour’s time #100FREEiPhone12 LIVE Event happening at this link - https://t.co/DwIzvtj8Z5
Retweet this tweet or the hashtag #100FREEiPhone12 starting now.
💯💯💯