Suggestion: Vocal Expression Profiles for Voice Mode


Currently, ChatGPT offers several voices with distinct timbres, but there is little control over how these voices express themselves during a conversation.

I would like to suggest the implementation of a Vocal Expression Profile system, allowing the user to choose not only the voice but also the interpretation style used.

The proposal does not consist of creating new voices, but of allowing existing voices to adopt different vocal behavior patterns.

Basic Profiles

Examples of profiles available to all users:

  • Neutral
  • Serious
  • Energetic
  • Professional
  • Robotic
  • Calm
  • Narrator

These profiles could automatically adjust characteristics such as:

  • Emotional intensity
  • Speech speed
  • Frequency of pauses
  • Intonation variation
  • Degree of formality
  • Expressiveness

Possible Expansions for Paid Plans

ChatGPT Plus

Additional profiles focused on emotions and personality:

  • Angry
  • Melancholic
  • Shy
  • Enthusiastic
  • Confident
  • Mysterious

ChatGPT Pro

Advanced profiles for specific uses:

  • Commentator
  • Professor
  • Sarcastic
  • Epic Narrator
  • Corporate Assistant
  • Boss
  • Scientist
  • Advisor

Benefits

  • Greater customization of the experience.

  • Better adaptation to different usage contexts.

  • Greater accessibility for users who prefer neutral or unemotional communication.

  • Additional differentiation between subscription plans.

  • Leveraging existing voices without the need to create dozens of new voices.

Conclusion

Allowing users to choose vocal expression profiles would make voice mode significantly more flexible and enjoyable for different audiences. The proposal leverages the existing voice infrastructure and adds a layer of personalization that can benefit both free users and subscribers to paid plans.

Hey @Jordo_JNS! This is a really thoughtful suggestion. Having expression profiles for existing voices, like calm, professional, energetic, or narrator, could make Voice Mode feel more personal without needing entirely new voices.

It would also help users match the speaking style to the situation, whether they want something neutral, more expressive, or easier to listen to. I can’t share a timeline right now, but I’ll pass this feedback along internally.

- Sunny