Inventor

AUDHKHASI KARTIK

US21 patents

⚠️ This page may combine multiple inventors who share the name “AUDHKHASI KARTIK”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

AUDHKHASI KARTIK

IBM

Implementing a whole sentence recurrent neural network language model for natural language processing

External word embedding neural network language models

End-to-end spoken language understanding without full transcripts

Detecting and recovering out-of-vocabulary words in voice-to-text transcription systems

Knowledge transfer between recurrent neural networks

Knowledge transfer between recurrent neural networks

Feature and feature variant reconstruction for recurrent model accuracy improvement in speech recognition

Aligning spike timing of models for maching learning

Leveraging unpaired text data for training end-to-end spoken language understanding systems

Soft-forgetting for connectionist temporal classification based automatic speech recognition

Implementing a whole sentence recurrent neural network language model for natural language processing

Recognition of out-of-vocabulary in direct acoustics-to-word speech recognition using acoustic word embedding

Training of model for processing sequence data

Multilingual intent recognition

Transliteration based data augmentation for training multilingual ASR acoustic models in low resource settings

GOOGLE LLC

Mixture model attention for flexible streaming and non-streaming automatic speech recognition

Mixture model attention for flexible streaming and non-streaming automatic speech recognition

Regularizing word segmentation

UNIV SOUTHERN CALIFORNIA

Noise-enhanced convolutional neural networks

Noise speed-ups in hidden markov models with applications to speech recognition

AUDHKHASI KARTIK

Automatic evaluation of spoken fluency