BBC-Esq

icon indicating a website link https://www.chintellalaw.com icon indicating an email [email protected]

icon company Blair Chintella, Esq. icon location Atlanta, Georgia metropolitan area icon location Lawyer teaching himself Python and AI as a hobby - Exhibit A.

Results 251 comments of BBC-Esq

Language support?

Translation support is planned within the next week.

Language support?

Should be completed in the latest release. Let me know if it doesn't work for some reason. https://github.com/BBC-Esq/Faster-Whisper-Transcriber/releases/tag/v5.1.0

[bug] Incorrect results for floating point numbers. `p.number_to_words(0.000001) != 'one hundred and six'`

I might be willing to help.

adopt modern bert please!

Some additional links to models just released: https://huggingface.co/answerdotai/ModernBERT-large https://huggingface.co/answerdotai/ModernBERT-base https://huggingface.co/blog/modernbert

Introduce Intel Arc GPU acceleration

Lack of interest, and I can't personally test since I don't have an ARC gpu.

Introduce Intel Arc GPU acceleration

I'll re-open this issue and start thinking about perhaps adding intel gpu support, but can't make any promises as to when. However, if/when I get the ball rolling, I'll continue...

evaluate code-specific embedding models

Also necessitates splitting .py or other programming language with langchain's code-specific splitting logic. Also add various programming file extensions to accepted list, use appropriate loaders, etc.

evaluate code-specific embedding models

https://code-representation-learning.github.io/codesage-v2.html

evaluate code-specific embedding models

https://huggingface.co/nomic-ai/nomic-embed-code

Issues with running examples text_to_audio_playback.py and gui_text_to_audio_playback.py circular import and webdataset.

Hmm...let me look into it.