β οΈ Note: The rates above reflect usage with the base model (gpt-3.5-turbo) and Azureβs regular voice engine.
Component | Cost per minute (Β’) |
---|---|
Transcription | 1 |
LLM | 1 |
Voice | 2 |
Platform | 2 |
Total | 6 |
Case | Multipliers (T, LLM, Voice, Platform) | Cost Breakdown (Β’) | Total (Β’/min) |
---|---|---|---|
Regular LLM + Azure Voice | 1, 1, 1, 1 | 1 + 1 + 2 + 2 | 6 |
Regular LLM + ElevenLabs Voice | 1, 1, 2.5, 1 | 1 + 1 + 5 + 2 | 9 |
Regular LLM + Cartesia Voice | 1, 1, 1.75, 1 | 1 + 1 + 3.5 + 2 | 7.5 |
Regular LLM + PlayHT Voice | 1, 1, 2.5, 1 | 1 + 1 + 5 + 2 | 9 |
Regular LLM + Voice (Your Keys) | 1, 1, 0, 1 | 1 + 1 + 0 + 2 | 4 |
o3-mini + ElevenLabs Voice | 1, 2, 2.5, 1 | 1 + 2 + 5 + 2 | 10 |
o3-mini + Azure Voice | 1, 2, 1, 1 | 1 + 2 + 2 + 2 | 7 |
OpenAI Realtime (4o-mini) | 0, 10, 0, 1 | 0 + 10 + 0 + 2 | 12 |
OpenAI Realtime (4o) | 0, 46, 0, 1 | 0 + 46 + 0 + 2 | 48 |
β± Voice usage is measured per second and prorated accordingly.
LLM Model | Multiplier | Cost per Query (Β’) |
---|---|---|
gpt-3.5-turbo | 1 | 1.5 |
o3-mini | 2 | 3.0 |
gpt-4 | 20 | 30 |
gpt-4o | 10 | 15 |