alphacep / alphacep/vosk-api

hardware dependend recognition problems: M$ Surface versus Logitech Webcam C925e

Open
#809 5 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
15.1k
Forks
1.8k
PR merge metrics
No merged PRs in 30d

Description

Different recording hardware seems to make a big difference in the overall accuracy.

On an AMD Ryzen5 with a Logitech C925e as input device at ~75% loudness level, the accuracy of the word "carola" with the current big german model is excellent, not perfect, but with 95% fine to use. This system has an nvidia card installed.

The same model on an i7 i915 chipset Surface Tablet with ~ the same word "carola" has a really bad recognition rate of ~30%. On the surface we don't have an nvidia device installed.

A tool to determine the perfect audio level of the microphone for recognition and a test tool for the hardware support would be nice. This could make debugging recognition mistakes a lot easier.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.