Billing of Speech-To-Text
Speech-to-Text Service supports the recognition of audio streams from specified users or all users in a TRTC room, converting them into text (Speech-To-Text, STT) and outputting the results.
Billing Instructions
If your application version is an RTC-Engine Monthly Package, you can unlock the Speech-to-Text capability. Other versions cannot unlock the Speech-to-Text capability.
Billing method: Daily post-paid billing. The fees generated the previous day are deducted at 10 AM each day.
Billing cycle: Billed daily. For detailed billing and invoice timing, please refer to the actual Billing Statement.
Pricing
The list price for the Speech-to-Text service is shown in the following table:
Billable Item | Unit Price (USD/Minute) | Supports 22 Languages |
Speech-to-Text Service | 0.02 | Supports Chinese, Traditional Chinese, English, Vietnamese, Japanese, Korean, Indonesian, Thai, Portuguese, Turkish, Arabic, Spanish, Hindi, French, Malay, Filipino, German, Italian, Russian, Swedish, Danish, Norwegian. |
Usage Calculation
Only the audio duration that starts to participate in Speech-to-Text is counted for usage statistics.
For multi-stream input from the host, the usage duration of each stream is summed up for billing.
Note:
When initiating a Speech-to-Text task for a TRTC room, the system assigns a virtual robot to act as an audience member. This robot subscribes to the required audio/video streams, and fees are calculated based on the duration of streams received by the robot. For details, refer to Billing of Audio and Video Duration.
Durations for each day are calculated in seconds for each application (SDKAppID) and then converted to minutes (rounded up to the nearest minute) for billing.
Billing Example
User A wants to use the Speech-to-Text service, so they need to purchase an RTC-Engine Monthly Package (Light/Standard/Pro version) to use this capability. Suppose they purchase the Engine Light version in February 2025 and enable the Speech-to-Text service. On February 1, the usage is 100 minutes.
On February 2, 2025, the Speech-to-Text fee that User A needs to pay is as follows:
Speech-to-Text fee = 100 (minutes) × 0.02 (USD) = 2 USD. This fee will be deducted around 10 AM on February 2, 2025.