Using stacked transformations for recognizing foreign accented speech.
Peter SmitMikko KurimoPublished in: ICASSP (2011)
Keyphrases
- speech recognition
- recognition engine
- speech signal
- speech synthesis
- endpoint detection
- audio visual
- automatic recognition
- dialogue system
- text to speech
- automatic speech recognition
- broadcast news
- noisy environments
- speech recognizer
- speech processing
- spoken language
- speaker recognition
- database
- multi lingual
- gaussian mixture model
- website
- feature selection
- vocal tract
- spontaneous speech
- data sets
- real time