Hello, my name is Florian Zimmermeister and I come from a small town in North Rhine-Westphalia, Germany.
I am passionate about speech recognition, both to provide more accessibility for people with disabilities and to make the value of data available to everyone by making it high quality transcribed, translated and searchable. For this, I train open models, work on training algorithms and datasets. I love working with the huggingface ecosystem and discussing about crazy ideas for the open source world.
Other open source contributions:
- First model with model card uploaded to huggingface hub (https://huggingface.co/aware-ai/bart-squadv2/commits/main)
- Working for SOTA ASR for german language since 2021 ( https://huggingface.co/primeline/whisper-large-v3-turbo-german , https://huggingface.co/primeline/whisper-large-v3-german , https://huggingface.co/primeline/whisper-tiny-german-1224 , https://huggingface.co/aware-ai/wav2vec2-xls-r-1b-5gram-german , https://huggingface.co/aware-ai/wav2vec2-large-xlsr-53-german-with-lm )
- As of today (21.03.2023), I am one of the top 20 most active listeners and top 50 most active speakers for the German Commonvoice dataset.
- As of today (11.05.2023) I am in the global Top 100 data collectors in the open assistant project.
- As of today (11.06.2023) I am in the global Top 50 data collectors in the open assistant project.
- November 2023 started working on the lorax inference server (https://github.com/predibase/lorax) until June 2024
- As of today (22.02.2024) I am in the global Top 5 data collectors in the argilla DIBT project.
- December 2024: 20x smaller ASR model matching performance of whisper medium for german speech recognition (https://huggingface.co/primeline/whisper-tiny-german-1224)
- End of 2024: More than 130k Downloads for the primeline/whisper-large-v3-german