Francisco Herrera

Sorry if this is already in the guide but I noticed that subjectively, LLMs perform better if prompted in English which is something many non-native English speakers overlook when prompting...

According to top comment here: https://www.reddit.com/r/vibecoding/comments/1mfyz71/are_people_actually_able_to_vibe_code_without/

in the future i might add this functionality.

whisper.cpp doesn't run

5

### zluda_trace logs (tarball/zip file) [log.zip](https://github.com/user-attachments/files/22710960/log.zip) ### Description Whisper.cpp when compiled with cuda support doesn't work with zluda. It says the CUDA driver version is insufficient for CUDA runtime version...

adding streaming/data pipeline/simultaneous processing of several transcription steps to speed up processing

1

From what i can see, whisper.cpp has several steps which are carried out one at a time before proceeding to the next: conversion with ffmpeg, vad detection, and then transcription,...

Clarify that selectedrows is used like an array

1

### Type of issue Missing information ### Description Clarify that selectedrows is used like an array. The sxample in the documentation doesn't make it very clear as someone new to...