[comp] Sähkötekniikan korkeakoulu / ELEC
Permanent URI for this collectionhttps://aaltodoc.aalto.fi/handle/123456789/77
Browse
Browsing [comp] Sähkötekniikan korkeakoulu / ELEC by Department "Signaalinkäsittelyn ja akustiikan laitos"
Now showing 1 - 3 of 3
- Results Per Page
- Sort Options
- Automatic Speech Recognition for Northern Sámi with comparison to other Uralic Languages
School of Electrical Engineering | A4 Artikkeli konferenssijulkaisussa(2016) Smit, Peter; Leinonen, Juho; Jokinen, Kristiina; Kurimo, MikkoSpeech technology applications for major languages are becoming widely available, but for many other languages there is no commercial interest in developing speech technology. As the lack of technology and applications will threaten the existence of these languages, it is important to study how to create speech recognizers with minimal effort and low resources. As a test case, we have developed a Large Vocabulary Continuous Speech Recognizer for Northern Sámi, an Finno-Ugric language that has little resources for speech technology available. Using only limited audio data, 2.5 hours, and the Northern Sámi Wikipedia for the language model we achieved 7.6% Letter Error Rate (LER). With a language model based on a higher quality language corpus we achieved 4.2% LER. To put this in perspective we also trained systems in other, better-resourced, Finno-Ugric languages (Finnish and Estonian) with the same amount of data and compared those to state-of-the-art systems in those languages. - Cooperative Game-theoretic Approach to Load Balancing in Smart Grids with Community Energy Storage
School of Electrical Engineering | A4 Artikkeli konferenssijulkaisussa(2015) Rajasekharan, Jayaprakash; Koivunen, VisaIn this paper, we propose a model for households to share energy from community energy storage (CES) such that both households and utility company benefit from CES. In addition to providing a range of ancillary grid services, CES can also be used for demand side management, to shave peaks and fill valleys in system load. We introduce a method stemming from consumer theory and cooperative game theory that uses CES to balance the load of an entire locality and manage household energy allocations respectively. Load balancing is derived as a geometric programming problem. Each household’s contribution to overall non-uniformity of the load profile is modeled using a characteristic function and Shapley values are used to allocate the amount and price of surplus energy stored in CES. The proposed method is able to perfectly balance the load while also making sure that each household is guaranteed a reduction in energy costs. - Morfessor 2.0: Toolkit for statistical morphological segmentation
School of Electrical Engineering | A4 Artikkeli konferenssijulkaisussa(2014) Smit, Peter; Virpioja, Sami; Grönroos, Stig-Arne; Kurimo, MikkoMorfessor is a family of probabilistic machine learning methods forfinding the morphological segmentation from raw text data. Recentdevelopments include the development of semi-supervised methods forutilizing annotated data. Morfessor 2.0 is a rewrite of the original,widely-used Morfessor 1.0 software, with well documented command-linetools and library interface. It includes algorithmic improvements and new features such as semi-supervised learning, online training, and integrated evaluation code.