# Emotion & sentiment from customer calls

> Multimodal models that combine what is said with how it is said, to flag frustrated customers and coach agents, on every call, not a sample.

Source: https://aibyos.com/research/emotion

Research · Voice & speech

# Emotion & sentiment from customer calls

Multimodal models that combine what is said with how it is said, to flag frustrated customers and coach agents, on every call, not a sample.

**100%**of calls analysed vs. manual sampling

**audio + text**multimodal signals

**minutes**from call end to insight

## Pipeline

1.  Diarise speakers and transcribe
2.  Extract acoustic features (pitch, pace, energy)
3.  Fine-tune a text + audio classifier on labelled calls
4.  Calibrate against QA-team judgements
5.  Dashboards and alerts in the contact-centre tools

Figures show the typical order of magnitude for this approach compared with calling a large general-purpose model. Actual results depend on the task and data; we measure them on your data during the baseline phase.

[Discuss a project like this](https://aibyos.com/contact)

## Typical tooling

-   pyannote
-   wav2vec 2.0
-   Transformers
-   Streamlit / BI

## Related

[All research](https://aibyos.com/research)[Sovereign GPU compute](https://aibyos.com/gpu)[Fine-tuning](https://aibyos.com/services/fine-tuning)