Mi-Go: Test Framework which uses YouTube as Data Source for Evaluating Speech Recognition Models like OpenAI's Whisper.
Tomasz WojnarJaroslaw HryszkoAdam RomanPublished in: CoRR (2023)
Keyphrases
- speech recognition
- probabilistic model
- acoustic models
- hidden markov models
- speech signal
- speech recognizer
- language model
- speech understanding
- speech synthesis
- speech processing
- automatic speech recognition
- noisy environments
- data sources
- audio visual speech recognition
- speech recognition technology
- speech recognition errors
- data mining
- speech recognition systems
- keyword spotting
- speaker diarization
- pattern recognition