TTS that sounds less fake

[amberpatch25]

This keeps coming up in conversation so I figured I'd start a thread.

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with tool calling.

Wondering if this is just where the whole space is heading now.

I care more about stability than squeezing the last possible tokens/sec out of it.

[quiettoken]

Can you fill in a bit more context? What model, what runtime, and roughly what context size?

[kernelpeak]

Ran a quick check on my box. On A6000 with llama.cpp, I saw basically the same behavior.

[bytegarden73]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[visionforge]

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

[cudascope]

Thanks, that helps. I ended up in about the same place using SGLang. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[batchshed]

I'd start with the simpler stack.

[vectorfield]

My first guess would be this: On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

[cobaltspark]

My first guess would be this: On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorou

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at NVFP4. Not glamorous advice, but repeatable baselines save a lot of time.

[kernelpeak]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Kimi and compare from there.