Share via

Pricing for Azure Voice Live API

Sankar Ramakrishnan, Prathap 20 Reputation points
2026-02-17T12:53:48.97+00:00

We are evaluating Azure Voice live API for our Contact Center use case, automating with AI. However, we could not find the latest pricing of Azure Voice live API - we want to use Pro version - use Azure speech, GPT 5.2 Chat (or suitable chat models). Could you provide latest pricing details.

Azure AI Speech
Azure AI Speech

An Azure service that integrates speech processing into apps and services.

0 comments
Answer accepted by question author
  1. Manas Mohanty 15,295 Reputation points Microsoft External Staff Moderator
    2026-02-17T15:58:06.01+00:00

    Hi Sankar Ramakrishnan, Prathap

    During Voice Live API interactions, you are charged

    1. Text tokens
    2. Audio input tokens
    3. Audio output tokens
    4. Cached audio and text context
    5. Any Azure Speech components used (standard or custom avatar, audio models etc.)

    Cost will differ based on scenarios.

    Scenario without custom text to speech model or avatars

    you can use speech pricing guide to estimate the cost incurred.

    Attached pricing in one of supported region (East US)

    User's image

    Example Scenario for usage in East US.
    Token Type Monthly Tokens
    Text Input 30,000,000
    Text Output 50,000,000
    Audio Input 100,000,000
    Audio Output 120,000,000

    Cached split (30%)

    • Cached Text Input: 9M
    • Cached Audio Input: 30M

    Cost Calculation

    • Text Input (non‑cached): 21M × $4.40 = $92.40
    • Cached Text Input: 9M × $1.375 = $12.38
    • Text Output: 50M × $17.60 = $880.00

    Total Text Cost: $984.78

    Please take a minute to accept this answer if you found it helpful.

    Thank you.

    1 person found this answer helpful.
    0 comments

1 additional answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Q&A Assist
    2026-02-17T12:54:31.8733333+00:00
    AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

    The pricing for the Azure Voice Live API is tiered based on the generative AI model used. For the Pro version, which includes models like GPT-5, the pricing details are as follows:

    • Voice Live Pro: This tier includes models such as gpt-realtime, gpt-4o, gpt-4.1, gpt-5, and gpt-5-chat.

    The specific pricing will take effect on July 1, 2025, and you will be charged based on the usage of text and audio with Azure Speech, as well as any custom models you may use. If you opt for custom speech, custom voice, or custom avatars, those will incur separate charges for model training and hosting.

    For more detailed pricing scenarios and examples, you can refer to the Azure documentation.

    References:

    0 comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.