Audio-to-Intent Using Acoustic-Textual Subword Representations from End-to-End ASR.
Pranay DighePrateeth NayakOggi RudovicErik MarchiXiaochuan NiuAhmed H. TewfikPublished in: CoRR (2022)
Keyphrases
- end to end
- broadcast news
- automatic speech recognition
- acoustic features
- multimedia
- speech recognition
- audio signal
- admission control
- high bandwidth
- wireless ad hoc networks
- speech signal
- ad hoc networks
- congestion control
- spoken document retrieval
- spontaneous speech
- rate allocation
- content delivery
- metadata
- internet protocol
- packet loss rate