zh217/torch-asg
Auto Segmentation Criterion (ASG) implemented in pytorch
This project offers a method for training deep learning models that process sequences, particularly useful in speech recognition. It takes raw audio data or similar sequential inputs and, after training, produces highly accurate transcriptions or classifications. This is designed for researchers and engineers developing advanced deep learning systems for sequence modeling.
No commits in the last 6 months.
Use this if you are building an end-to-end speech recognition system or similar sequence-to-sequence model and need an alternative loss function to CTC that offers different theoretical advantages and potentially better integration with WFSTs.
Not ideal if you are looking for a pre-trained, ready-to-use speech recognition solution or if you are not working with deep learning models in PyTorch.
Stars
51
Forks
9
Language
C++
License
GPL-3.0
Category
Last pushed
Oct 01, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/zh217/torch-asg"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
TensorSpeech/TensorFlowASR
:zap: TensorFlowASR: Almost State-of-the-art Automatic Speech Recognition in Tensorflow 2....
dangvansam/viet-asr
VietASR - Vietnamese Automatic Speech Recognition
wenet-e2e/wenet
Production First and Production Ready End-to-End Speech Recognition Toolkit
xinjli/allosaurus
Allosaurus is a pretrained universal phone recognizer for more than 2000 languages
srvk/eesen
The official repository of the Eesen project