Latency is too high even higher than 30sec
# support
h
Hi Team Vapi , I need quick support, the Latency for voice assistants are getting higher and higher day by day . The product is no more suitable to demo even . Customers are just running away from demo , by looking at such unstable product. Please help .... Here is the call id - 89567d00-804e-4f6a-99dd-018be4cfb15a
v
message has been deleted
You can continue this conversation by mentioning me in the message.
@Hirekarigar I'm really sorry to hear that you're experiencing issues with latency. We understand how important it is to have a reliable and fast-performing voice assistant for your demos and customer interactions. To address this issue promptly, could you please provide me with more details about the circumstances under which you're experiencing high latency? For example, the specific operations or functions that are slow, the time of day when this occurs, and any error messages you might be receiving. This information will help us to diagnose and resolve the problem more efficiently.
n
hey, that call id is <1s, could you share a call id with high latency?
h
Here is another call id 00ad1de3-127c-4e45-80df-e9f0c2b7b8e1
n
you can see the in the dashboard the turn-by-turn latency
Copy code
šŸ”µ 16:00:21:037 [user INFO] Turn Latency: 4431ms (Transcriber: 119ms, Endpointing 100ms, Model 3569ms, Voice: 601ms)
seems model for your use case is just slow
h
Model speed depends upon the number of requests you can make in a mins....thats controlled by groq and hence vapi . So can you please increase that ?
2 Views