Free on-line speech recogniser based on Kaldi ASR toolkit producing word posterior lattices.
Ondrej PlátekFilip JurcícekPublished in: SIGDIAL Conference (2014)
Keyphrases
- automatic speech recognition
- word error rate
- speech recognition
- spontaneous speech
- word recognition
- speech recognition systems
- speech recognizer
- speech recognizers
- speech signal
- recognition errors
- conversational speech
- spoken document retrieval
- handwriting recognition
- language model
- error rate
- broadcast news
- hidden markov models
- spoken language
- noisy environments
- prosodic features
- human machine interaction
- speech corpus
- spoken words
- speech sounds
- speech retrieval
- keyword spotting
- lexical features
- acoustic features
- co occurrence
- probability distribution
- automatic transcription
- posterior distribution
- speech synthesis
- n gram
- probabilistic model
- spoken documents
- keywords
- text to speech
- posterior probability
- bayesian framework
- pattern recognition
- high quality
- vocal tract
- handwritten words
- english text
- word level
- speaker verification