CONF
Schnell_MLSLP-18_2018/IDIAP
A Neural Model to Predict Parameters for a Generalized Command Response Model of Intonation
Schnell, Bastian
Garner, Philip N.
EXTERNAL
https://publications.idiap.ch/attachments/papers/2018/Schnell_MLSLP-18_2018.pdf
PUBLIC
https://publications.idiap.ch/index.php/publications/showcite/Schnell_INTERSPEECH2018_2018
Related documents
MLSLP-18 Proceedings
Hyderabad
2018
Satelite workshop of Interspeech '18
https://sites.google.com/view/mlslp/home
URL
This abstract summarizes our paper accepted in the main
conference with the same title.
CONF
Schnell_INTERSPEECH2018_2018/IDIAP
A Neural Model to Predict Parameters for a Generalized Command Response Model of Intonation
Schnell, Bastian
Garner, Philip N.
EXTERNAL
https://publications.idiap.ch/attachments/papers/2018/Schnell_INTERSPEECH2018_2018.pdf
PUBLIC
Proc. Interspeech 2018
2018
3147-3151
/10.21437/interspeech.2018-1904
doi
The Generalised Command Response (GCR) model is a time-local model of intonation that has been shown to lend itself to (cross-language) transfer of emphasis. In order to generalise the model to longer prosodic sequences, we show that it can be driven by a recurrent neural network emulating a spiking neural network. We show that a loss function for error backpropagation can be formulated analogously to that of the Spike Pattern Association Neuron (SPAN) method for spiking networks. The resulting system is able to generate prosody comparable to a state-of-the-art deep neural network implementation, but potentially retaining the transfer capabilities of the GCR model.