Martin @discuss.tchncs.de Dedicated to high-quality #SpeechToText #transcription #STT
Personally interested in #Python #OpenAI #WebDevelopment #FireFox
Website: https://www.speech-to-text.cloud/
This comment shows why I like Lemmy more than Reddit. Nuanced, acknowledging when the other person has a point without just yelling at each other.
Custom Linux Distribution just for Gaming
Bazzite is a custom image built upon Fedora Atomic Desktops that brings the best of Linux gaming to all of your devices - including your favorite handheld.
Bazzite comes ready to rock with Steam and Lutris pre-installed, HDR support, BORE CPU scheduler for smooth and responsive gameplay, and numerous community-developed tools for your gaming needs.
A font with an LLM embedded
You type "Once upon a time!!!!!!!!!!" and those exclamation marks are rendered to show the LLM generated text, using a tiny 30MB model
via https://simonwillison.net/2024/Jun/23/llama-ttf/
They use stem cells derived from human skin
FinalSpark builds processor from 16 organoid brain cells
FinalSpark has unveiled Neuroplatform, a pioneering approach to
Large Language Models made from cells incoming...
Kind of ironic how they use an AI generated article image
I made a Speech-to-text transcription website built with Python, FastAPI, and Traefik
Upload an .mp3 and download your text transcript immediately. No subscription and no account required. Super-easy Speech to Text conversion.
Hello Entrepreneur community,
I'm Martin, and I recently launched a new tool that I'd like to share with the community. It's a website for speech-to-text transcription that allows users to upload an audio file and receive a transcription of the spoken text.
The platform is built using Python, FastAPI, and Traefik, and offers the following features:
- Speech recognition using Whisper
- High accuracy transcriptions
- HTTP routing and SSL termination handled by Traefik
- Web framework provided by FastAPI
I believe this tool can be useful for a variety of applications, such as transcribing podcasts, interviews, or any other audio content. The accuracy of the transcriptions is quite high (WER score of 4.5), and I am constantly working on improving it even further.
I am looking for feedback on the platform, as well as any potential use cases or applications that you can think of. I am also interested in collaborations or partnerships, so please don't hesitate to reach out if you are interested in working together.
If this tool has piqued your interest and you would like to learn more, I encourage you to check it out and leave a comment with your thoughts. Your feedback will help me improve the platform moving forward.
Thank you for taking the time to check out my project.