🎙️ Mumble AI
STT: Groq · Whisper Large V3 Turbo
AI: Groq · GPT-OSS-20B
Mode: Auto
Ready
Tone / Style

Transcript Sandbox

Raw Transcript

—

Polished Output

—

Hold Space or F8 to speak. Double-tap Space for Hands-free. Esc cancels.

Latency
—
Speed
—
Zero-edit
—
Duration
—
ASR
—
Est. Cost
$0.0000
Quick Samples

History

Search and review past transcriptions, latencies, and estimated costs.

Multi-Model AI Hub

Configure Speech-to-Text and AI Enhancement independently with transparent per-unit pricing.

Speech-to-Text (STT)

Transcribes microphone voice streams into raw text

0 models

AI Enhancement (LLM)

Polishes grammar, cleans syntax, and formats output

0 models

Live Cost Estimator

Simulate real-time infrastructure costs for your chosen model pair.

Selected: Groq Whisper + Groq GPT-OSS-20B
Estimated STT Cost
$0.000154
Estimated LLM Cost
$0.000330
Total Estimated Cost
$0.000484

Estimated cost based on published provider rates. Actual charges depend on usage volume and provider billing tiers.

Usage & Cost Analytics

Track your audio minutes, token usage, monthly quotas, and separated BYOK vs Platform infrastructure costs.

Monthly Quota Tier: Free
BYOK Active User API keys are 100% exempt from platform quotas
Mumble Audio Minutes 0.0 / 60.0 min (0%)
Mumble AI Requests 0 / 500 reqs (0%)
TODAY
0.0 min
Audio Spoken
THIS MONTH (TOTAL)
0.0 min
Monthly Spoken
PLATFORM COST
$0.0000
Mumble-Managed Infrastructure
BYOK VOLUME
$0.0000
Direct User Key Spend

Breakdown by Provider

Provider Requests Audio Platform Cost BYOK Cost
No usage recorded yet.

Breakdown by Model

Model Type Requests Usage Est. Cost
No usage recorded yet.

Recent Usage & Audit History

Honest telemetry log with labeled cost estimates, BYOK attribution, and failure diagnostics.

Time Type Provider / Model Audio / Tokens Latency Key Source Cost Status
No events recorded.

Personal Dictionary

Names, technical terms, and spoken phrase mappings that should be spelled accurately every time.

Saved Entries

Voice & Text Snippets

Say or type a trigger phrase to instantly expand templates, links, or code snippets.

Saved Snippets

Meeting Intelligence

Multi-speaker transcription with structured executive notes.

Sprint planning
Stephen, Nathalie, Mikel
02:15

Notes

Generate notes when you want a summary of this conversation.

Command Mode

Paste a passage, then say how it should change.

Source Text

Refined Result

Would you like to discuss our recent software deployment later this month?

Platform Administration

Manage user accounts, review tenant usage, and inspect system telemetry.

TOTAL USERS
0
Registered Accounts
DICTATION VOLUME
0
Total Dictations Completed
TOKEN WORKLOAD
0
Total AI Tokens
SYSTEM HEALTH
Healthy
Uptime: 0h 0m

User Accounts & Access Control

Active accounts can authenticate and submit voice dictation.

User / Email Role Status Registered Usage Summary Account Actions
Loading accounts...

Provider Infrastructure & Cost Separation

Month-to-date breakdown separating Mumble platform expenses from user BYOK spend.

Provider Total Requests Audio Minutes Total Tokens Platform Cost BYOK Volume Failures
Loading provider telemetry...

Active AI Configuration

Server-wide defaults and provider API key availability.

Default STT Provider-
Default STT Model-
Default LLM Provider-
Default LLM Model-
Default Tone-
Groq Key Configured-
Gemini Key Configured-
OpenAI Key Configured-

Host Telemetry & Infrastructure

Runtime environment details and database status.

Node.js Runtime-
Operating System-
Process Memory (RSS)-
Database StorageSQLite (WAL Mode enabled)
Environment Mode-
Catalog STT Models-
Catalog LLM Models-

System Failure & Reliability Log

Recent provider or processing errors with timestamps and root-cause error messages.

Time User Operation Provider / Model Latency Credential Source Error Diagnostic
No recent system failures recorded.