Sesame vs Kintsugi

Comparing the features of Sesame to Kintsugi

Feature
Sesame
Kintsugi

Capability Features

AI Fairness and Bias Mitigation
Consistent Personality
Context Awareness
Conversational Dynamics
Conversational Speech Generation
Dataset Size
1 million hours
Digital Health Focus
Emotional Intelligence
Evaluation Suite
Focus on Depression and Anxiety
DepressionAnxiety
Healthcare Support
Inpatient and Outpatient Use
InpatientOutpatient
Mental Health Screening Tool
Model Sizes
Tiny: 1B backbone, 100M decoderSmall: 3B backbone, 250M decoderMedium: 8B backbone, 300M decoder
Multiple Speaker Handling
Objective Metrics
Word Error RateSpeaker SimilarityHomograph DisambiguationPronunciation Consistency
Partial Multilingual Support Planned
Planned for 20+ languages
Pronunciation Correction
Remote Team Support
Sequence Length
2048
Short Speech Clip Analysis
Single-Stage Model
Subjective Metrics
Comparative Mean Opinion Score
Text and Audio Input
TextAudio
Training Epochs
5
Unbiased Data Acquisition
Voice Biomarker Detection

Integration Features

GitHub Release
LLama Architecture Backbone
Mimi Split-RVQ Tokenizer
Pega Integration

Limitation Features

Cannot Model Conversation Structure
English Language Dominance
Memory Bottleneck in Training
No File Format Disclosure
No Free Tier Mentioned
No Pre-trained Language Model Use
No Public API Listed
No Public Pricing
Real-Time Generation Delay
RVQ time-to-first-audio scales poorly

Pricing Features

Free Preview
Open Source
Apache 2.0