A Reddit user is encountering problematic and unrealistic audio outputs when using AI models, specifically mentioning issues with adult-themed scenes. Despite attempts to refine prompts by removing descriptive words related to liquids and sounds, the AI continues to generate bizarre noises like slurping or slide whistles. The user has experimented with different models and LoRAs but found the issue persists, suggesting it may be inherent to the base model. AI
RANK_REASON User is complaining about audio output from a model, not a new release or significant event.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →