SADiLaR Language Resource Repository
Not a member yet
536 research outputs found
Sort by
MELFA (Mobile E-Learning for Africa)
No full textEmbedded TTS on mobile phones providing speech output of full texts for various types of applications
Lwazi Sepedi ASR corpus
No full textComplete audio recordings and orthographic transcriptions used for Lwazi speech recognition systems
University of Stellenbosch Xhosa-English Speech-speech translation
No full textEnglish -> Xhosa & Xhosa -> English speech to speech translation in a naval context (ASR speech input & TTS output) (Pc-based?
Autshumato TMS
No full textTerminology Management System. Web application used by Terminologists and Administrators to capture, edit and export terminology
Sesotho Genre Classification Corpus
No full textContains training and testing data for Genre Classification for Sesotho
Autshumato English-Sesotho sa Leboa Parallel Corpora
No full textParallel corpora aligned on sentence level through a combination of automatic and manual alignment techniques. The parallel corpora were obtained from the SA government domain
Lwazi Xitsonga Pronunciation Dictionary
No full textGeneral phonemic pronunciations for frequently occurring words in SA languages. Dictionaries were developed to be practically usable for speech technology systems, rather than phonetically accurate. Audio samples of all phonemes included. A letter-to-sound rule set for predicting the pronunciations of generic words included. (Separate entry describes rule sets.
Lwazi Siswati ASR corpus
No full textComplete audio recordings and orthographic transcriptions used for Lwazi speech recognition systems
(SRP-Type A Research) Ability-based technology interventions
No full textResearch conducted to investigate how the use of multiple modalities can improve training for persons based on their abilities, language, literacy. One possible output modality to be used is speech synthesis (if the person does not have the ability to see), whilst speech recognition as input modality can be used for a person that has the ability to talk and not the ability to physically interact with the PC. (A technology demonstrator is currently available in English & South African Sign Language
Open Secretary (vSep 2009)
No full textA telephone-based automated attendant for call routing that uses ASR/DTMF (input) used in-house by the HLT RG at Meraka Institute, CSIR. Routing is based on ASR name recognition. Calls are routed to an office phone or cell-phone (upon no-answer)