Speech collage: code-switched audio generation by collaging monolingual corpora.
Amir HusseinDorsa ZeinaliOndrej KlejchMatthew WiesnerBrian YanShammur Absar ChowdhuryAhmed M. AliShinji WatanabeSanjeev KhudanpurPublished in: CoRR (2023)
Keyphrases
- audio visual
- audio stream
- broadcast news
- cepstral features
- speaker identification
- audio signals
- text to speech
- parallel corpus
- emotion recognition
- chinese english
- natural language processing
- question answering
- statistical machine translation
- speech processing
- machine translation
- audio files
- audio recordings
- digital audio
- speech recognition
- speech music discrimination
- query expansion
- source code
- linear predictive coding
- multimedia
- prosodic features
- domain specific
- audio features
- cross lingual
- speech synthesis
- spoken documents
- speech signal
- information retrieval
- automatic transcription
- multi modal
- acoustic signals
- audio video
- multi stream
- visual information
- ad hoc retrieval
- spoken document retrieval
- speech retrieval
- word alignment
- information extraction
- acoustic features