Audio docs

[tukwila]

Summary

Details

• [ ]

Provide audio benchmark practice guide based on mock-server.

Test Plan

Related Issues

• Resolves #

---

• "I certify that all code in this PR is my own, except as noted below."

Use of AI

• Includes AI-assisted code completion
• Includes code generated by an AI application
• Includes AI-generated tests (NOTE: AI written tests should have a docstring that includes ## WRITTEN BY AI ##)

[jaredoconnell]

Please fill in the necessary components of the PR template. Especially the certification that the work is your own, and the AI section if applicable.

[sjmonson]

Needs python -m mdformat docs/examples/practice_on_audio_benchmark.md to make lintier happy.

[sjmonson]

Since everything below refers to 127.0.0.1, just bind to the one address:

Suggested change

	guidellm mock-server --host 0.0.0.0 --port 8080
	guidellm mock-server --host 127.0.0.1--port 8080

[sjmonson]

Suggested change

	git clone https://github.com/vllm-project/guidellm.git
	cd guidellm
	pip install guidellm
	pip install guidellm

[sjmonson]

Since you install gTTS above, why not use it to generate a dataset here? Something like:

from gtts import gTTS
from faker import Faker
import wave

...

for i in range(count):
    prompt = faker.text(max_nb_chars=100)
    tts = gTTS(prompt)
    tts.save(audio_filename)

    ...

I did not do any verification on this so it may not work.