Hey think about the bright side of voice AI -- there are new opportunities like AI companions, AI assistant, not only the antiquated contact methodologies.
If you are referring to the LLM used in the demo, it's a simple GPT. If you are referring to audio data, there are some (not a lot) public datasets, although be careful of the license of the dataset. To get more data, you might want to build a studio to collect from contracted voice actors, or you can purchase from other sources.
We are dedicated to preventing that from happening. Spam calling and identity theft are key areas we will build guardrails around. Feel free to let us know if you think of any other case.
Good point, note token. benchmarking is a great tool to show differentiation. BTW, apart from what we think is important ourselves (latency, mean opinion score, etc), would you mind sharing what you want to see in such a benchmark? One key metric I like to keep an eye on is the end conversion rate of using the product, but that's very use-case specific.
Nice catch on the working -- customer is indeed more accurate than user.
For outage handling: we strive to keep up 99.9 plus up time, and in the case of a dropped call, the agent would hang up if using phone, and might have different error handling in web depending on how customer handles it.
I totally get that clicking in Web UI is super convenient in many scenarios, and I think GUI and voice can co-exist and create synergy. Suppose AI voice agent can solve your problem, cater to your needs, and interact like a human. In that case, I believe it would be super helpful in many scenarios (like others mentioned, waiting on line for 40 minutes is a pretty bad experience). There are also new opportunities in voice like AI companion, AI assistant, etc that we see are starting to emerge.
Could you elaborate a bit on "my speech engine tied to a specific model provider"? Sorry, I might be lacking some context on what you are referring to here.
This can be handled with function calling and other features in LLM. We support the input signal of closing the call, and you can have your rule-based (timer) system or LLM-based end call functionality and use that to hang up.
Yes, I agree there are problems with LLM (hallucinations, persona, etc), and that's exciting because that means room for improvement and opportunities. I know many people who are working hard in that field trying to make LLM converse better.
For example
- "hallucinations / LLMs confabulate": techniques like RAG can help
- "Language is so smooth that you often need the expertise to catch them in it", fine-tuning and prompt engineering can help
We handle the audio bytes in / out, and also connect to your user's server for response. We handle the interaction and decide when to listen and when to talk, and send live updates to our users. When a response is needed, we ask for it and get it from our user.