When I activate GPU acceleration my speech is not converted to text

[Cosnet]

I am on Linux Mint and I installed from the applications manager Speech Note. I downloaded and trying to use the English (FasterWhisper Large-v3)/EN model. If I check to use in Engine options of the Speech to Text tab and I select: Auto (Cuda, 0, NVIDIA GeForce RTX 5070 Ti) and i speak a line of text. It seems to process but no text appears. If I shut it off then it works fine. My system is quite powerful AMD Ryzen 7 9800X3D processor, 128 GB of Ram, 4TB SSD. Just as a note and maybe this is what broke it. I used Google Gemini to ask it about the best speech to text model to use and it recommended the model I am using but it also said to activate GPU so I ran this command in the Terminal to install the nvidia addon.. "flatpak install net.mkiol.SpeechNote.Addon.nvidia"
Can someone help me troubleshoot this?
Also I am running Linux Mint version 22.3

[Cosnet]

I decided to download another speech to text model and selected the English WhisperCpp Large-v3 model and turned on gpu acceleration and it works. So it looks like it is related to to the FasterWhisper model.

[mkiol]

Hi. Thanks for reporting.

I just checked, and the ctranslate2 library (used for FasterWhisper) is not compiled with support for your GPU. The highest supported NVIDIA architecture is Hopper, but your card is a generation newer. That is why GPU acceleration is not working. I will try to fix this in the new version.

WhisperCpp (included in flatpak package) supports Blackwell and is a solid and highly efficient engine. FasterWhisper does not offer you higher speed, so I recommend you to continue to use WhisperCpp.