# MELD

MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversation

Repository: https://github.com/declare-lab/MELD
Canonical: https://ross.abutalabs.com/products/meld
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: emotion-recognition, sentiment-analysis, multimodal-sentiment-analysis, multimodal-interactions, dialogue-systems, conversational-ai, chatbot, personality-traits, personality-profiling, emotion, dialogue, emotion-detection, multimodal-emotion-recognition, emotion-recognition-in-conversation
Last push: 2026-05-17T04:23:40+00:00

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 82, release rhythm 35, longevity 100
- inputs: {"age_days": 2895, "days_push": 108, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1079, forks 234 (observed 2026-08-28T04:03:30.348011+00:00)

## What it is
A research repository from declare-lab containing implementations of neural architectures (DialogueRNN, DialogueGCN, COSMIC, bcLSTM) for emotion recognition in conversations, along with the MELD multimodal dataset. It serves as both a benchmark dataset and reference codebase for conversational emotion recognition research.

## Use cases
- recognize emotions in conversations
- train emotion recognition models on the MELD dataset
- benchmark conversational emotion recognition architectures
- run COSMIC for state-of-the-art ERC
- extract emotion causes from dialogue
- research multimodal sentiment analysis in conversations

## When to choose
- you need the MELD dataset or baselines for emotion recognition in conversation research
- you want reference implementations of DialogueRNN, DialogueGCN, or COSMIC
- you are doing academic work on conversational emotion or sentiment analysis

## When to avoid
- you need a production-ready emotion detection API or service
- you want a maintained library with stable APIs rather than research code
- your task is not conversation-level emotion recognition

## Facets
- artifact type: dataset
- maturity: maintenance
- function: nlp, machine-learning, deep-learning
- domain: machine-learning, artificial-intelligence
- platform: python
- tags: emotion-recognition, conversational-ai, sentiment-analysis, pytorch, multimodal, dialogue-systems, meld-dataset, natural-language-processing

## Member repositories
- declare-lab/MELD (main) score 69
- declare-lab/conv-emotion (examples) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:30.348011+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:31:40.943240+00:00, confidence not recorded.
  - readme: https://github.com/declare-lab/MELD (fetched 2026-08-28T04:03:30.348011+00:00, sha 4a57d35245dd)
- Data as of 2026-08-30T08:39:29.467469+00:00.