Katrin Kirchhoff

APPLICATION CREATION ENVIRONMENT USING MULTI-MODALITY INTERFACE OPTIONS

Joint ASR and speaker error correction post-processing

Reward weighting alignment of large language models

Constrained generative framework application development environment

Alignment model as decoding-time heuristic

Multi-domain fine-tuning for contextual adapters of automatic speech recognition

Augmenting automated speech recognition neural networks with scalable vocabularies

Speaker diarization error correction

Application creation environment using multi- modality interface options

Customizable latency for automatic speech recognition

DYNAMIC LANGUAGE MODEL UPDATES WITH BOOSTING

Infusing knowledge graphs into automatic speech recognition

Dynamic language model updates with boosting

Removing bias from automatic speech recognition models using internal language model estimates

DIALOG-DRIVEN APPLICATIONS SUPPORTING ALTERNATIVE VOCAL INPUT STYLES

REDACTING PORTIONS OF TEXT TRANSCRIPTIONS GENERATED FROM INVERSE TEXT NORMALIZATION

Dynamically selectable automated speech recognition using a custom vocabulary

Managing access to sensitive data in transcriptions

DOMAIN SPECIALTY INSTRUCTION GENERATION FOR TEXT ANALYSIS TASKS

Dialog-driven applications supporting alternative vocal input styles

Personalized batch and streaming speech-to-text transcription of audio

LARGE LANGUAGE MODELS PROVIDING EVIDENCE MAPPINGS FOR GENERATED OUTPUT

DOMAIN ENTITY EXTRACTION FOR PERFORMING TEXT ANALYSIS TASKS

AUTOMATED EVALUATION OF EVIDENCE MAPPING MODELS

Redacting portions of text transcriptions generated from inverse text normalization

MEDICAL CONVERSATION SUMMARIZATION STYLE INTELLIGENCE

Domain-specific parameter pre-fixes for tuning automatic speech recognition

Contextual biasing of neural language models using metadata from a natural language understanding component and embedded recent history

Multimodal based punctuation and/or casing prediction

Contextual natural language understanding for conversational agents

Multi-modal spoken language understanding systems

Model training system for custom speech-to-text models

Extending sensitive data tagging without reannotating training data