c2b_default

Robopsychologie

JKU - MA Psychologie

JKU - MA Psychologie

111
0.0 (0)

Kartei Details

Karten 111
Sprache English
Kategorie Technik
Stufe Universität
Erstellt / Aktualisiert 21.06.2020 / 25.10.2020
Weblink
https://card2brain.ch/box/20200621_robopsychologie
Einbinden
<iframe src="https://card2brain.ch/box/20200621_robopsychologie/embed" width="780" height="150" scrolling="no" frameborder="0"></iframe>
Lernen

Stapel: Tactile, Auditive, Visual Perception

WaveNet by Google DeepMind

  • Duplex‘ naturally sounding voice was developed using DeepMind's WaveNet technology

  • WaveNet directly models the raw waveform of the audio signal. It is a fully convolutional neural network (CNN). Input sequences are real waveforms recorded from human speakers. After training, the network is sampled to generate synthetic utterances

  • Able to generate speech which mimics any human voice and which sounds more natural than the best existing Text-to-Speech systems.

How humanlike should a virtual assistant sound?

  • Is it important for users to be able to distinguish between human and machine (e.g. on the phone)? Would you like to know?

  • Are artificial voices, which sound very humanlike, sometimes creepy?

  • Would you prefer to talk to a realistically human sounding bot or one that clearly sounds like a machine?

  • Are these preferences perhaps context-dependent?

  • Which visual image of a virtual conversation partner does a more or less human-like voice actually evoke?

Research at the LIT Robopsychology Lab:
User Expectations of Robot Appearance Induced by Different Robot Voices

Results

  • Human-likeness of the drawn robots was generally high across all conditions.

  • Some features appeared in almost all drawings regardless of the voice (e.g., head, eyes).

  • Other features were significantly more prevalent in voice conditions characterized by low human-likeness (wheels) or high human-likeness (e.g., nose).

„Female“ over-representation in voice assistants

  • Most companies that produce automated voices hold auditions for voice actors and collect recordings of them speaking. Then they invite focus groups to rate the voices on how well they convey certain attributes: e.g., warmth, friendliness, competence

  • Some studies suggest that female synthetic voices are preferred (voicebot.ai, 2019) as they are perceived as warmer compared to male voices (Karl MacDorman, Indiana University).

  • Other studies revealed the opposite:

    • Results indicated that female human speech was rated as preferable to female synthetic speech,

      and that male synthetic speech was rated as preferable to female synthetic speech

      (Mullenix et al., 2003).

    • Male voices are perceived as more intelligent (Clifford Nass, Stanford University)

Q – The first genderless voice

  • This first gender-neutral artificial voice was created to reduce gender bias in AI assistants.

  • Between ??? and ??? Hz (= gender-neutral range according to research)

  • Between 145-175 Hz (= gender-neutral range according to research)

  • Voice was refined after surveying 4,600 people

  • Collaboration: Copenhagen Pride, Virtue, Equal AI, Koalition Interactive & thirtysoundsgood

Lernen