The still increasing number of transistors per chip oered by Moore’s law, together with the Post-Dennard scaling era shifted the performance gain from frequency increase to multi-core processing. Consequently, the support of parallel execution of applications is becoming mandatory. Furthermore, the need for ecient parallel models and languages is more critical for the embedded domain, due to power consumption and memory constraints, among others. This work focuses on parallelizing an embedded speaker recognition application, which is a biometric technique for identication. While a lot of work has been done for speech recognition, fewer eorts have focused on recognizing who the speaker is. In this paper, we analyze two implementations for speaker recognition applications (SRA), namely dataow and shared memory programming models. More precisely, we use Process Networks (PNs) as a dataow representation, which is an intuitive way to design streaming applications. We use the language “C for Process Networks” for the dataow implementation and OpenMP for the shared memory one. For two dierent target platforms, we compared two implementations using OpenMP (exploring data-level parallelism only and with pipelining) against a dataow-based compiled implementation that allows for functional optimization. Despite faster communication over shared memory, we show that the dataow model is superior in terms of performance (up to twice as fast).