Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili.
Christiaan JacobsNathanaël Carraz RakotonirinaEverlyn Asiko ChimotoBruce A. BassettHerman KamperPublished in: CoRR (2023)
Keyphrases
- speech recognition
- speech recognition systems
- automatic speech recognition
- speech recognizers
- speech sounds
- speech recognizer
- spontaneous speech
- word error rate
- spoken language
- speech signal
- compound words
- english text
- noisy environments
- prosodic features
- acoustic features
- speech synthesis
- word recognition
- recognition errors
- conversational speech
- acoustic models
- speech segments
- multi lingual
- broadcast news
- vocal tract
- co occurrence
- speech corpus
- false positives
- automatic transcription
- spoken words
- text input
- text to speech
- speaker identification
- language model
- lexical features
- out of vocabulary
- speech retrieval
- dialogue system
- language independent
- vector space