Upload agent–farmer call recordings in any audio format. Transcribes and separates speaker turns (Urdu, Punjabi, Siraiki and more), scores each call against SOPs you define, tags sentiment, and tracks agent performance over time.
Speaker turns are AI-inferred from conversation content, not true audio diarization — treat "agent" / "farmer" labels as a best estimate, not a verified transcript split.
Define the criteria a call is judged against. Each gets scored 0–max by the QA model, with reasoning.
Aggregated from every call scored and saved on this device.
| Agent | Calls | Avg score | Positive | Neutral | Negative |
|---|---|---|---|---|---|
| No data yet. | |||||
Calls scored, average score, and sentiment mix across every call saved on this device (not just this session).
Tracked client-side for this browser session only (resets on reload) against each provider's daily limits, so you can see how much runway is left. Transcription shows Groq's meter or your self-hosted server's unlimited status depending on what's selected in Settings; speaker-separation and scoring always run on Groq. "Unconfirmed ceiling" models aren't documented publicly — the number shown is a conservative estimate, not a guarantee.
Choose how audio gets turned into text. Speaker-separation and QA scoring always use Groq (see the API key below), regardless of this choice.
Used for every scoring / speaker-separation call, and for transcription too if Groq is selected above. Saved in this browser's local storage after you enter it once, so you won't be asked again next time you open this file here. Free key at console.groq.com.
Agents, SOPs, and call results are saved automatically to this browser's local storage — no account, no setup. This means the data lives only on this device, in this browser, and clearing your browser data will erase it. Use the backup file below to move data to another device, share it with a teammate, or keep a safety copy.