Compact acoustic model for embedded implementation
نویسندگان
چکیده
An acoustic model for an embedded speech recognition system must exhibit two desirable features; ability to minimize performance degradation in recognition while solving the memory problem under limited system resources. To cope with the challenges, we introduce the state-clustered tied-mixture (SCTM) HMM as an acoustic model optimization. The proposed SCTM modeling shows a significant improvement in recognition performance as well as a solution to sparse training data problem. Moreover, the state weight quantizing method achieves a drastic reduction in model size. In this paper, we describe the acoustic model optimization procedure for embedded speech recognition system and corresponding performance evaluation results.
منابع مشابه
Compact and Efficient Active Vibro-acoustic Control of a Smart Plate Structure
An effective wide band active control law through one kind of the Dynamic Vibration Absorber (DVA) is proposed and studied in this paper. With the help of mechanical impedance method, active DVA control law is formulated based on the passive mechanical model. The electrical DVA can generate multi-mode active damping to the structure. The host structure is an aluminum plate and acceleration sign...
متن کاملImplementation of an adaptive burst DQPSK receiver over shallow water acoustic channel
In an environment such as underwater channel where placing test equipments are difficult to handle, it is much practical to have hardware simulators to examine suitably designed transceivers (transmitter/receiver). The simulators of this kind will then allow researchers to observe their intentions and carry out repetitive tests to find suitable digital coding/decoding algorithms. In this p...
متن کاملAchieving a reliable compact acoustic model for embedded speech recognition system with high confusion frequency model handling
An acoustic model for an embedded speech recognition system must exhibit two desirable features; the ability to minimize the performance degradation in recognition, while solving the memory problem under the constraint of limited system resources. Moreover, for general speech recognition tasks, context dependent models such as state-clustered tri-phones are used to guarantee the high recognitio...
متن کاملParallel implementation of underwater acoustic wave propagation using beamtracing method on graphical processing unit
The mathematical modeling of the acoustic wave propagation in seawater is the basis for realizing goals such as, underwater communication, seabed mapping, advanced fishing, oil and gas exploration, marine meteorology, positioning and explore the unknown targets within the water. However, due to the existence of various physical phenomena in the water environment and the various conditions gover...
متن کاملCompact Acoustic Models for Embedded Speech Recognition
Speech recognition applications are known to require a significant amount of resources. However, embedded speech recognition only authorizes few KB of memory, few MIPS and small a amount of training data. In order to fit the resource constraints of embedded applications, an approach based on a semi-continuous HMM system using stateindependent acoustic modelling is proposed. A transformation is ...
متن کامل