Comparing open source ASR toolkits on Italian children speech.
Piero CosiMauro NicolaoGiulio PaciGiacomo SommavillaFabio TesserPublished in: WOCCI (2014)
Keyphrases
- open source
- automatic speech recognition
- speech recognition
- speech signal
- speech recognition technology
- hearing impaired
- word error rate
- spontaneous speech
- open source software
- noisy environments
- hidden markov models
- source code
- broadcast news
- pilot study
- speech synthesis
- speech corpus
- autistic children
- spoken words
- case study
- conversational speech
- pattern recognition
- speech recognizer
- web services
- language model
- handwriting recognition
- speech retrieval
- text to speech
- spoken language
- data sets
- audio visual
- spoken document retrieval
- speaker identification
- core components
- language acquisition
- error rate
- speech sounds
- multi modal
- multi user environment
- neural network