Questionário vinculado ao guia · Médio Nível

Questionário de IA em Call Center Speech Analytics

Teste o reconhecimento de fala, a diarização e a análise de sentimentos em pesquisas em call centers, incluindo os limites e o escopo de cada estudo citado.

Caminhos de guia relacionadosIA em análise de fala de call center
Pergunta 1 de 8

What output can automatic speech recognition provide for speech analytics?

Visualize cada pergunta8 perguntas

As respostas e explicações aparecem conforme você responde ao teste.

  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?
    • Traduz a chamada para outro idioma
    • Segments the recording into speaker turns and assigns speaker labels
    • Mede o volume da chamada
    • Criptografa a gravação
  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. Qual recurso acústico ajuda a detectar o estado emocional de um cliente além das palavras usadas?
    • Tamanho do arquivo da gravação
    • O ID de funcionário do agente
    • A operadora de telefonia
    • Tom, ritmo e volume (prosódia)
  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