- Client
- Real estate agency, Kazakhstan
- Industry
- Real estate
Every sales call scored against the checklist
Speech-to-text and a language model review calls that a manager used to sample by hand.
The problem
The sales team of a new-build real estate agency was on the phone all day. The head of sales could listen to only a small sample, so most mistakes in scripts and objection handling went unnoticed.
What I built
- Call recordings are transcribed with Yandex SpeechKit.
- A language model scores each transcript against the team's QA checklist.
- Queueing, retries and logging: a failed recording is visible and is processed again.
The result
The pipeline ran in production and replaced a large share of manual call review. The head of sales reviewed scores instead of hours of audio.
For the same agency I built an outbound AI voice agent. Packages and prices for such systems are on the call and document analysis page.
Solution behind it
Pipelines with LLMs
Call and document analysis with LLMs
Pipelines that transcribe sales calls and score them against your checklist, or turn invoices and contracts into validated structured data.
- Python
- FastAPI
- OpenAI API
- Yandex SpeechKit
from$500