-
oliverguhr/fullstop-punctuation-multilang-large
Token Classification • Updated • 368k • 154 -
oliverguhr/fullstop-dutch-punctuation-prediction
Token Classification • Updated • 94 • 3 -
oliverguhr/fullstop-punctuation-multilingual-sonar-base
Token Classification • Updated • 656 • 2 -
kredor/punctuate-all
Token Classification • Updated • 505k • 18
Oliver Guhr
oliverguhr
AI & ML interests
Voice Interfaces, Robotics, Deep Learning
Recent Activity
liked
a model
4 days ago
ssary/XLM-RoBERTa-German-sentiment
liked
a model
30 days ago
openGPT-X/Teuken-7B-instruct-research-v0.4
liked
a model
30 days ago
jinaai/jina-embeddings-v2-base-de
Organizations
Collections
2
-
Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling
Paper • 2311.00430 • Published • 57 -
SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
Paper • 2307.01952 • Published • 82 -
Language Modeling Is Compression
Paper • 2309.10668 • Published • 83 -
Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Paper • 2311.00871 • Published • 2
Papers
1
models
12
oliverguhr/wav2vec2-large-xlsr-53-german-cv13
Automatic Speech Recognition
•
Updated
•
62
•
1
oliverguhr/spelling-correction-english-base
Text2Text Generation
•
Updated
•
95.5k
•
74
oliverguhr/fullstop-punctuation-multilang-large
Token Classification
•
Updated
•
368k
•
154
oliverguhr/wav2vec2-large-xlsr-53-german-cv9
Automatic Speech Recognition
•
Updated
•
110
•
1
oliverguhr/spelling-correction-multilingual-base
Text2Text Generation
•
Updated
•
109
•
9
oliverguhr/spelling-correction-german-base
Text2Text Generation
•
Updated
•
36.9k
•
15
oliverguhr/fullstop-dutch-punctuation-prediction
Token Classification
•
Updated
•
94
•
3
oliverguhr/fullstop-dutch-sonar-punctuation-prediction
Token Classification
•
Updated
•
1.56k
•
6
oliverguhr/fullstop-punctuation-multilingual-base
Token Classification
•
Updated
•
1.54k
•
6
oliverguhr/fullstop-punctuation-multilingual-sonar-base
Token Classification
•
Updated
•
656
•
2
datasets
None public yet