Help with Audio / Speech: what are you seeing?

[signalcraft14]

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

The more I work with local setups, the more it seems like router policy decides everything downstream.

Curious whether that lines up with what other people are seeing.

[dockkite69]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at KV cache pressure. Once you know the plain path works, then add the faster path back in piece by piece.

[kernelhouse70]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at KV cache pressure. On

Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[localpatch78]

Ran a quick check on my box. On 7900 XTX with PyTorch, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.

[amberkite]

Good point. I don't think your conclusion is far off. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[sparrowkite53]

Can you fill in a bit more context? Are you testing this on GB200 or something else?

[flintriver50]

Thanks, that helps. I ended up in about the same place using Ollama.

[signalpeak]

My first guess would be this: On A6000, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.

[sanescope]

Interesting, I hadn't tried that. That matches what I saw on dual 3090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[cudapoint]

Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly. I expected the model choice to matter more than it did.