AI Text Emotion Detector

Classify the emotions expressed in English text: joy, sadness, anger, fear, surprise, neutral and more.

AI Private · runs on your device Free
Runs locally in your browser AI processing happens on your device. Your files are not uploaded for inference.
AI model
…
Download size
… ·
Device
Checking browser support…

Classifies the emotions expressed in English text (28 labels). It is not a psychological assessment of the writer.

Discover the emotional tone of a text

The Text Emotion Detector uses a RoBERTa model trained on Google's GoEmotions dataset to score 28 emotion labels: joy, gratitude, love, curiosity, surprise, sadness, anger, fear, disappointment, neutral and more.

Use cases

  • Understanding the tone of customer reviews and support tickets
  • Gauging the mood of social media posts
  • Testing which emotions marketing copy evokes
  • NLP teaching and research

Important note

The result describes emotions expressed in the wording; it's not a diagnosis or a judgment about the writer. Irony, context and culture can change meaning. The model is trained on English text. Your text is processed on your device and never sent anywhere.

How to use AI Text Emotion Detector

  1. Paste your English text.
  2. The model scores the emotion labels.
  3. Review the top-scoring emotions.
  4. Treat the results as a model estimate.

Why use this tool?

Analyse the emotional tone of text for free without compromising privacy.

FAQ

Can it tell what someone really feels?

No. It only classifies the wording. Irony, context and culture can change meaning; don't use it to judge people.

Can it analyse non-English text?

The model is trained on English data, so results for other languages aren't reliable.

Does it do positive/negative sentiment?

It offers 28 more detailed emotion labels; you can interpret positive and negative sentiment from them.

Related tools