On automatic phonetic transcription quality: lower word error rates do not guarantee better transcriptions.
Judith M. KessensHelmer StrikPublished in: Comput. Speech Lang. (2004)
Keyphrases
- automatic transcription
- spoken document retrieval
- high quality
- quality assessment
- spontaneous speech
- error bounds
- speech recognition
- spoken term detection
- error rate
- speech recognition systems
- co occurrence
- semi automatic
- information retrieval
- broadcast news
- maximum error
- spoken documents
- fully automatic
- word sense disambiguation
- word segmentation