Exploiting Hidden Representations from a DNN-based Speech Recogniser for Speech Intelligibility Prediction in Hearing-impaired Listeners.
Zehai TuNing MaJon BarkerPublished in: CoRR (2022)
Keyphrases
- hearing impaired
- speech recognition
- linear prediction
- speech synthesis
- speech signal
- prediction accuracy
- audio visual
- speaker identification
- vocal tract
- endpoint detection
- automatic speech recognition
- dialogue system
- prediction error
- spoken language
- prediction model
- spontaneous speech
- audio stream
- bit rate
- automatic speech recognition systems