Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

tavus.io

Post reply on HN
Print view View original

41–50 of 50 posts

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#41

Post by code_brian »

BizarroLand wrote: post #37
butlike wrote: post #36

Earlier quoted context omitted.

Am I not allowed to cut you off if you're ramble-y and incoherent?

Its rude if you're a human, and entirely unacceptable if you are a computer.

The one thing that really surprised me, the thing I learned that's affected my conversational abilities the most: turn taking in conversation is a negotiation: there are no set rules. There are protocols: - bids - holds / stays - implications (semantic / prosodic)

But then the actual flow of the conversation is deeply semantic in the best conversations, and the rules are very much a "dance" or a negotiation between partners.

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#42

Post by sippeangelo »

allan_s wrote: post #23

How does it compare with https://github.com/KoljaB/RealtimeVoiceChat , which is absent of the benchmark ?

That's not a turn-taking model, it's just a silence detection Python script based on whatever text comes out of Whisper...

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#43

Post by code_brian »

ljoshua wrote: post #20

Hey @code_brian, would Tavus make the conversational audio model available outside of the PALs and video models? Seems like this could be a great use case for voice-only agents as well.

You can reach out to our sales team. You can chat with our AI SDR here, and they will review it and reach out. https://www.tavus.io/demo

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#44

Post by code_brian »

sigmoid10 wrote: post #17
ttul wrote: post #9

I tried talking to Claude today. What a nightmare. It constantly interrupts you. I don’t mind if Claude wants to spend ten seconds thinking about its reply, but at least let ME finish my thought. Without decent turn-taking, the AI seems impolite and it’s just an icky experience. I hope tech like this gets widely distributed soon because there are so many situations in which I would love to talk with a model. If only…

Anthropic doesn't have any realtime multimodal audio models available, they just use STT and TTS models slapped on top of Claude. So they are currently the worst provider if you actually want to use voice communication.

It's unfortunate though, because Anthropic LLMs and ecosystem is the best IMHO. Tavus (we) and Anthropic should form a partnership.

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#45

Post by BizarroLand »

code_brian wrote: post #41
BizarroLand wrote: post #37

Earlier quoted context omitted.

Its rude if you're a human, and entirely unacceptable if you are a computer.

The one thing that really surprised me, the thing I learned that's affected my conversational abilities the most: turn taking in conversation is a negotiation: there are no set rules. There are protocols: - bids - holds / stays - implications (semantic / prosodic) But then the actual flow of the conversation is deeply semantic in the best conversations, and the rules are very much a "dance" or a negotiation between p…

That's an interesting way to think about it, I like that.

It also implies that being the person who has something to say but is unable to get into the conversation due to following the conversational semantics is akin to going to a dance in your nice clothes but not being able to find a dance partner.

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#48

Post by code_brian »

BizarroLand wrote: post #45
code_brian wrote: post #41

Earlier quoted context omitted.

The one thing that really surprised me, the thing I learned that's affected my conversational abilities the most: turn taking in conversation is a negotiation: there are no set rules. There are protocols: - bids - holds / stays - implications (semantic / prosodic) But then the actual flow of the conversation is deeply semantic in the best conversations, and the rules are very much a "dance" or a negotiation between p…

That's an interesting way to think about it, I like that. It also implies that being the person who has something to say but is unable to get into the conversation due to following the conversational semantics is akin to going to a dance in your nice clothes but not being able to find a dance partner.

Yeah, I can relate to that. Maybe it's also because you are too shy to ask someone to dance. I think I learned that lesson: just ask, and be unafraid to fail. Things tend to work themselves out. Much of this is experimentation. I think our models need to be open to that: which is one cool thing about Sparrow-1: it's a meta-in-context learner. This means that when it try's and fails, or you try and fail, it learns at runtime to adapt.

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#49

Post by sigmoid10 »

code_brian wrote: post #44
sigmoid10 wrote: post #17

Earlier quoted context omitted.

Anthropic doesn't have any realtime multimodal audio models available, they just use STT and TTS models slapped on top of Claude. So they are currently the worst provider if you actually want to use voice communication.

It's unfortunate though, because Anthropic LLMs and ecosystem is the best IMHO. Tavus (we) and Anthropic should form a partnership.

I think Anthropic currently has a slight edge for coding, but this is changing constantly with every new model. For business applications, where tool calling and multi-modality matter a lot, OpenAI is and always has been superior. Only recently Google started to put some small dents in their moat. OpenAI also has the best platform, but less because it is good and more because Google and Anthropic are truly dismal in every regard when it comes to devx. I also feel like Google has accrued an edge in hard-core science, but that is just a personal feeling and I haven't seen any hard data on this yet.

Re: Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR

#50

Post by arkobel »

Have you compared with Krisp-TT models? https://krisp.ai/blog/krisp-turn-taking-v2-voice-ai-viva-sdk... Krisp LLC also shares an End-of-Turn Test dataset. Did you test your model on that? https://huggingface.co/datasets/Krisp-AI/turn-taking-test-v1

And can you share some information about the model size and FLOPS?

Post reply on HN