Rehber bağlantılı sınav · Orta Seviye

Çağrı Merkezinde Yapay Zeka Konuşma Analitiği Testi

Alıntı yapılan her çalışmanın sınırları ve kapsamı da dahil olmak üzere, çağrı merkezi araştırmasında konuşma tanıma, günlük tutma ve duygu analizini test edin.

İlgili kılavuz yollarıÇağrı Merkezinde Yapay Zeka Konuşma Analitiği
Soru 1 arasında 8

What output can automatic speech recognition provide for speech analytics?

Her soruyu önizleyin8 sorular

Testi çözdüğünüzde cevaplar ve açıklamalar görünür.

  1. What output can automatic speech recognition provide for speech analytics?
    • A machine-generated transcript for downstream text analysis
    • A guaranteed speaker identity
    • A final diagnosis of the caller’s emotional state
    • A payment authorization
  2. What does speaker diarization do in a call recording?
    • Aramayı başka bir dile çevirir
    • Segments the recording into speaker turns and assigns speaker labels
    • Çağrının ne kadar yüksek olduğunu ölçer
    • Kaydı şifreler
  3. In IBM’s two-stream call-center emotion-recognition study, which information streams are combined?
    • Video and keystroke timing
    • Customer purchase history and account balance
    • Acoustic and lexical information
    • Only transcript punctuation
  4. Hangi akustik özellik, kullanılan kelimelerin ötesinde müşterinin duygusal durumunu tespit etmeye yardımcı olur?
    • Kaydın dosya boyutu
    • Temsilcinin çalışan kimliği
    • Telefon taşıyıcısı
    • Perde, tempo ve hacim (prozodi)
  5. In a two-stream call-center emotion system, what do acoustic features contribute?
    • Evidence from vocal delivery, such as how speech sounds
    • The caller’s verified identity
    • The account’s payment history
    • A guaranteed diagnosis of the caller’s emotion
  6. What task did Vidrascu and Devillers study for customer-service phone calls?
    • Predicting credit scores from call recordings
    • Speaker segmentation and emotion-related analysis of conversational turns
    • Generating live agent sales recommendations
    • Automating legal decisions about customer accounts
  7. Why might a call-center sentiment system combine lexical and acoustic features?
    • Acoustic features replace the transcript in every task
    • Lexical features reveal the phone carrier
    • The words and the way they are spoken provide complementary signals
    • Combining them guarantees the caller’s true emotion is known
  8. What does word-error rate compare in speech recognition evaluation?
    • Speaker identity against a customer account
    • Emotion labels against call duration
    • Call volume against agent schedule
    • Substitutions, deletions and insertions against the words in a reference transcript