Product Details
Czech SpeechDat(E) Database
Created: 2000
speech processing
The Czech SpeechDat(E) Database (EasternEuropean Speech Databases for Creation of Voice Driven Teleservices)comprises 1052 Czech speakers (526 males, 526 females) recorded overthe Czech fixed telephone network. This database is partitioned into 6CD's. The speech databases made within the SpeechDat(E) project werevalidated by SPEX, the Netherlands, to assess their compliance withthe SpeechDat(E) format and content specifications.The speech files are stored as sequences of 8-bit, 8kHz A-law speechfiles and are not compressed, according to the specifications ofSpeechDat(E). Each prompt utterance is stored within a separate fileand has an accompanying ASCII SAM label file. The database can be ordered from European Language Resources Associationunder catalog number S0094:
http://www.elda.fr/cata/speech/S0094.html