Bark Mood Decoder is built on published bioacoustics. It reads the emotion in the sound, not words, and it is honest about how sure it is.
Bark Mood Decoder measures real acoustic signals: pitch, loudness, how fast barks repeat, and how tonal versus harsh they are. It reads emotion, not words, because dogs do not use them.
Dogs bark differently depending on what they feel, and it's measurable.
Machine learning sorts bark context to 43% across six situations, well above the 18% expected by chance.
We read arousal and valence the same way behavioral researchers do.
Our honesty rule: we show confidence, the second likeliest read, and say unclear when a sound is ambiguous. The most reliable signal is your own dog's trend over time.
Stays on your phone. Lets Bark Mood Decoder build a baseline unique to your dog.
You've watched it build over your last reads. Pro sharpens every read against your own dog and keeps the full story.