Recurrent neural networks (RNNs) are state-of-the-art in voice awareness/understanding and speech recognition. On-device computation of RNNs on low-power mobile and wearable devices would be key to applications such as zero-latency voice-based human-machine interfaces. Here we present CHIPMUNK, a small (<1mm2) hardware accelerator for Long-Short Term Memory RNNs in UMC 65 nm technology capable to operate at a measured peak efficiency up to 3.08Gop/s/mW at 1.24 mW peak power. To implement big RNN models without incurring in huge memory transfer overhead, multiple CHIPMUNK engines can cooperate to form a single systolic array. In this way, the Chipmunk architecture in a 75 tiles configuration can achieve real-time phoneme extraction on a demanding RNN topology proposed in [1], consuming less than 13 mW of average power.

Chipmunk: A systolically scalable 0.9 mm2, 3.08Gop/s/mW @ 1.2 mW accelerator for near-sensor recurrent neural network inference

Conti, Francesco
;
Benini, Luca
2018

Abstract

Recurrent neural networks (RNNs) are state-of-the-art in voice awareness/understanding and speech recognition. On-device computation of RNNs on low-power mobile and wearable devices would be key to applications such as zero-latency voice-based human-machine interfaces. Here we present CHIPMUNK, a small (<1mm2) hardware accelerator for Long-Short Term Memory RNNs in UMC 65 nm technology capable to operate at a measured peak efficiency up to 3.08Gop/s/mW at 1.24 mW peak power. To implement big RNN models without incurring in huge memory transfer overhead, multiple CHIPMUNK engines can cooperate to form a single systolic array. In this way, the Chipmunk architecture in a 75 tiles configuration can achieve real-time phoneme extraction on a demanding RNN topology proposed in [1], consuming less than 13 mW of average power.
2018 IEEE Custom Integrated Circuits Conference, CICC 2018
1
4
Conti, Francesco; Cavigelli, Lukas; Paulin, Gianna; Susmelj, Igor; Benini, Luca
File in questo prodotto:
File Dimensione Formato  
Binder4.pdf

accesso aperto

Tipo: Postprint
Licenza: Licenza per accesso libero gratuito
Dimensione 1.19 MB
Formato Adobe PDF
1.19 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/652926
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 27
  • ???jsp.display-item.citation.isi??? 20
social impact