Running
Whisper Base & Small Β· 60-Clip Listening Gallery
π§
Self-contained Base/Small S10 predictions on 60 audio clips
LAION eV
non-profit
AI & ML interests
open multi-modal foundation models and datasets for their creation; scaling laws, model evaluation; fully local, sovereign model deployment, personalized assistants and open local agentic systems
Recent Activity
View all activity
Running
Humaneness Ears audio listening atlas
π§
Humaneness Ears Base/Medium audio listening atlas
Running
Humaneness Voice Small Acting Evaluation
π
Explore and compare voice acting model performance across stages
Running
Whisper Emotion, Voice and Burst Gallery
π
Whisper Tiny/Base audio scores and burst timing
Running
Humaneness Voice Small Studies
π
Explore performance data and audio samples of Humaneness Voice Small
Running
Humaneness Voice Small DPO LoRA Acting Comparison
π
Running
Can a voice-acting model produce screams and shrieks?
π¬
Generate onβdemand scream and shriek voice clips
Running
MOSS Voice-Acting β Technical Report
π
Generate expressive speech from scripted directions
Running
Voice Pairs β Listening Demo
π§
Running
Vocal Burst Segments β Samples
π«
Explore labeled vocal burst audio samples
Running
Burst LoRAs trained on Gemini labels vs the shipped ones
π
Compare audio burst adapters and view detection results
Running
Gemini Burst Verification
π
Explore annotated vocal burst samples with audio playback
Running
Gemini as a vocal-burst annotator
π§
Explore Geminiβs burst annotations on drama audio clips
Configuration error
Burst Guidance Listening
π
Running
Scream and Shriek β what the burst adapters actually produce
π±
Explore generated scream &β―shriek audio at varying strengths
Running
MOSS Vocal Burst Recipes
π£
Listen to vocal burst samples and see detection results
Running
Synthesised Vocal Bursts β does it work?
π
Listen to synthetic vocal bursts and assess their labels
Running
Emotion Continuity β 20 Transitions
π
Rate emotional crossfades in AIβgenerated speech clips
Running
MOSS Quality Adapter Listening Test
π§
Rate and compare synthetic speech with different quality adapters
Running
Crossfade v2 β Listening Study
π
Compare emotional voice transitions and rate smoothness
Running
Emotional Transitions β Listening Study
π
Rate emotional voice transitions by listening to audio clips
Running
MOSS Voice-Acting β Emotion LoRA Merge-Weight Ablation
π
Adjust emotion intensity in synthetic speech with LoRA weights
Running
MOSS Voice-Acting β Turning the Emotion Up at Inference Time
π
Adjust emotional tone in generated speech
Running
What Makes a Generated Voice Sound Emotional
π
A controlled 4-factor study on the MOSS voice-acting model