Akten, Memo and Grierson, Mick. 2016. 'Real-time interactive sequence generation and control with Recurrent Neural Network ensembles'. In: Recurrent Neural Networks Symposium, NIPS 2016. Barcelona, Spain 8 Dec 2016. [Conference or Workshop Item]

Preview

Text rnn_ensemble_nips.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (663kB) | Preview

Abstract or Description

Recurrent Neural Networks (RNN), particularly Long Short Term Memory (LSTM) RNNs, are a popular and very successful method for learning and generating sequences. However, current generative RNN techniques do not allow real-time interactive control of the sequence generation process, thus aren’t well suited for live creative expression. We propose a method of real-time continuous control and ‘steering’ of sequence generation using an ensemble of RNNs and dynamically altering the mixture weights of the models. We demonstrate the method using character based LSTM networks and a gestural interface allowing users to ‘conduct’ the generation of text.

Item Type:	Conference or Workshop Item (Poster)
Keywords:	recurrent neural network, long short term memory, interactive, natural language processing, text generation, human computer interaction
Departments, Centres and Research Units:	Computing
Dates:	Date Event 1 November 2016 Accepted
Event Location:	Barcelona, Spain
Date range:	8 Dec 2016
Item ID:	19352
Date Deposited:	16 Jan 2017 13:05
Last Modified:	29 Apr 2020 16:21
URI:	https://research.gold.ac.uk/id/eprint/19352

View statistics for this item...

Edit Record (login required)