01 / ON-PREMISES
Own hardware
Deploy transcription, analysis and reporting inside your environment.
How we implement it
- Set up local STT and GPU servers
- Deploy text models for analysis
- Integrate with local call recording storage
- Technology options
- Local STT, speaker separation, a local text model, storage and BI; GPUs sized from benchmarks.
- Data & dependencies
- Audio and analysis remain local when the recording source and models are local too. Mask agreed fields and define deletion periods.
- Budget components
- Models, servers, sample annotation and BI integration; then GPUs, storage, quality evaluation and maintenance.
- Operations & support
- The support agreement assigns ownership of hardware, OS, applications, updates and incident response. Recovery is tested before launch.