[var.media_title;onformat=retitle]

由 S Majumdar 著作 · 2021 · 被引用 69 次 — Citrinet is deep residual neural model which uses 1D time-channel separable convolutions combined with sub-word encoding and squeeze-and- ...

由 X Wu 著作 · 2022 · 被引用 1 次 — Abstract:Citrinet is an end-to-end convolutional Connectionist Temporal Classification (CTC) based automatic speech recognition (ASR) model.

Citrinet is a deep residual convolutional neural network architecture that is optimized for Automatic Speech Recognition tasks. There are many variants of the ...

2023年9月5日 — The NVIDIA TAO Toolkit eliminates the time-consuming process of building and fine-tuning DNNs from scratch for IVA applications.

# This config contains the default values for training a Citrinet model with CTC loss and BPE-based vocabulary. # Default learning ...

2024年1月18日 — Citrinet-CTC model is an autoregressive variant of Citrinet model [1] for Automatic Speech Recognition which uses CTC loss/decoding instead ...

7 天前 — Streaming Citrinet-1024 model is a non-autoregressive, streaming variant of Citrinet model [1] for Automatic Speech Recognition which uses ...

Citrinet is deep residual neural model which uses 1D time-channel separable convolutions combined with sub-word encoding and squeeze-and-excitation. The ...