Emotion in an airplane announcement
Listen to “The airplane is almost full” and inspect how one recording receives several emotion scores alongside a tired speaking style.
Listen to the recording
A boarding announcement describes a fact. Can a service also represent several qualities in how the speaker delivers it?
Take 1 · 2.94 seconds
Transcript: “The airplane is almost full.”
| Output | Label | Score |
|---|---|---|
| emotions | worried | 0.992 |
| emotions | sad | 0.946 |
| emotions | disappointed | 0.915 |
| styles | tired | 0.982 |
Reading the scores
CREMA-D labels the performance as sad. The model also returned worried and disappointed. Each score is independent, so reducing this clip to a single winning label discards part of the output.
These acted English recordings were selected from the training split for demonstration. The scores describe audible delivery. Use a separate held-out set to measure accuracy.
Recording credits and licenses: CREMA-D. The public sample catalog includes each source file, audio hash and saved output.
Run this example
curl --fail --location https://oruk.ai/samples/emotional/1038_TAI_SAD_XX.wav -o take.wav
curl --fail-with-body https://speech-api.oruk.ai/v1/audio/analysis \
-H "Authorization: Bearer $ORUK_API_KEY" \
-F file=@take.wav -F model=oruk-resonanceThe example uses a key with speech API access. The table shows the saved run. Open the matching SDK quickstart.
Try it on your own audio.
The demo needs no account. The trial needs a card.