No project description provided
Project description
Introduction
This project implements the optimization techniques proposed in Improving RNN Transducer Modeling for End-to-End Speech Recognition to reduce the memory consumption for computing transducer loss.
During the implementation, we use torchaudio as a reference. We ensure that the transducer loss computed using this project is identical to the one computed using torchaudio when giving the same inputs.
Installation
You can install it via pip
:
pip install optimized_transducer
Installation FQA
What operating systems are supported.
It has been tested on Ubuntu 18.04. It should work on macOS and other unixes systems. Also, it may also work on Windows, though it is not tested.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
optimized_transducer-0.9.1.tar.gz
(40.5 kB
view hashes)
Close
Hashes for optimized_transducer-0.9.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8270e3211e7537179e04fc731e4bf28c8d7c5edf01cb6eb03f8bfe424569ca40 |
|
MD5 | 8523bb8bceb8e9df524084166a823fef |
|
BLAKE2b-256 | 4f6c0f8d5ee42424d8b5e3461f9e4c10bdf0812e673f9e52001ecd8651de1407 |