Open Settings
Whisper, speaker labels, summaries and chat each have their own model selection.
COMPATIBLE MODELS
Transcription uses Whisper, not a proprietary ASR model. You can switch the installed model in Settings and download supported files through the in-app catalog. Meeting content is processed locally; license activation and model downloads use the internet. Summaries, chat and search use local GGUF models.
Whisper, speaker labels, summaries and chat each have their own model selection.
The catalog shows variants suitable for your GPU video memory. Downloading models requires internet access.
The selected model is used when the next task for that function begins.
COMPATIBLE MODELS
Promic Speech AI uses Whisper for transcription and local third-party models for other tasks. In Settings, choose an installed model for transcription, speaker labels, summaries and chat; the in-app catalog offers compatible files for download.
Whisper Large V3 Turbo (default), Large V3, Medium Q8, Medium English Q8 and Small.
Medium English Q8 is for English audio only.
Qwen3.5 4B, 9B, 35B-A3B and 122B-A10B; Qwen3.8 27B; Gemma 4 E4B, 12B, 26B-A4B and 31B; GPT-OSS 20B.
The catalog identifies the role of each variant; for example, Qwen3.5 4B is offered for speaker labels.
Qwen3-Embedding-0.6B Q8_0 and BGE Reranker v2 m3 Q8_0.
Supporting models for indexing, search and result ranking.
These are third-party models downloaded through the in-app catalog. Availability depends on the installed version, NVIDIA GPU, video memory, edition and file variant. Compatibility of external files outside the catalog is not guaranteed.
FAQ
No. Transcription uses local Whisper models, and you can choose among the files in the in-app catalog.
The application lets you select installed files and add catalog entries, but compatibility is established for the listed variants. Other files may fail to load or may not suit the selected task.
No. A workstation without an NVIDIA GPU, or set to consultation only, can access the archive and review but cannot process new meetings.
Professional: US$20 per month per workstation, with a free 15-day trial. The application has no transcription-hour quota or per-token charge; practical capacity depends on the GPU, model and storage. It requires 64-bit Windows and an NVIDIA RTX GPU with at least 8 GB of VRAM. Enterprise and Government editions are offered by proposal, with architecture and capabilities defined for each deployment.
Features