Turkish Speech Recognition Database ---- Sentences (Desktop)-201 Speakers (King-ASR-159)
All the audio files were manually transcribed and annotated
- Speech Corpus
This database was collected in Turkey by Speechocean’s project team. This database is one of our databases of Speech Data ----Desktop Project (SDD) which contains the database collections in 30 languages presently.
It contains the voices of 201 different native speakers who were balanced distributed in age(mainly 16 – 30,31 – 45,46 – 60), Gender (104 males, 97 females)and regional accents (for the details, please see the technical document). The script was specially designed to provide material for both training and testing of many classes of speech recognizers, each speaker will be recorded in a quiet office environment and 300 phonetically rich sentences which was randomly selected from a pool of sentences specially designed.
The speech data are stored as sequences of 44.1 kHz, 16 bit and uncompressed. A pronunciation lexicon with a phonemic transcription in SAMPA is also included. The pure recording hour will be 153.55 hours.
- Speech Recognizer
- Speech Recognition Applications
- Speech Recognition System
The database was made for the Training and testing purpose of speech recognition systems for Turkish speech ASR applications.