Originally published at: https://developer.nvidia.com/blog/controlled-adaption-of-speech-recognition-models-to-new-domains/
Using adapters for parameter-efficient training reduces the effects of catastrophic forgetting of general speech recognition.
jwitsoe
1
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Speech Recognition: Customizing Models to Your Domain Using Transfer Learning | 0 | 408 | November 9, 2021 | |
An Easy Introduction to Speech AI | 1 | 533 | June 25, 2022 | |
New on NGC: One Click Deploy, AI Models for Speech and Computer Vision, and More | 0 | 381 | May 4, 2022 | |
Dynamic Memory Compression | 1 | 16 | January 24, 2025 | |
Announcing NVIDIA NeMo: Fast Development of Speech and Language Models | 0 | 416 | August 25, 2020 | |
How to Build Domain Specific Automatic Speech Recognition Models on GPUs | 0 | 448 | August 25, 2020 | |
Neural Modules for Fast Development of Speech and Language Models | 0 | 340 | August 25, 2020 | |
Deep Speech: Accurate Speech Recognition with GPU-Accelerated Deep Learning | 6 | 553 | February 22, 2016 | |
Enhancing Customer Experience in Telecom with NVIDIA Customized Speech AI | 0 | 290 | May 30, 2023 | |
Adding External Knowledge and Controllability to Language Models with Megatron-CNTRL | 1 | 447 | October 7, 2020 |