> ## Documentation Index
> Fetch the complete documentation index at: https://openrouter.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# STTResponse

STT response containing transcribed text and optional usage statistics

## Example Usage

```typescript theme={null}
import { STTResponse } from "@openrouter/sdk/models";

let value: STTResponse = {
  text: "Hello, this is a test of OpenAI speech-to-text transcription.",
};
```

## Fields

| Field      | Type                                             | Required             | Description                                                                            | Example                                                                                                                              |
| ---------- | ------------------------------------------------ | -------------------- | -------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ |
| `duration` | *number*                                         | :heavy\_minus\_sign: | Duration of the input audio in seconds, present when response\_format is verbose\_json | 9.2                                                                                                                                  |
| `language` | *string*                                         | :heavy\_minus\_sign: | Detected or forced language, present when response\_format is verbose\_json            | english                                                                                                                              |
| `segments` | [models.STTSegment](../models/sttsegment.mdx)\[] | :heavy\_minus\_sign: | Timestamped transcript segments, present when response\_format is verbose\_json        |                                                                                                                                      |
| `task`     | *string*                                         | :heavy\_minus\_sign: | The task performed, present when response\_format is verbose\_json                     | transcribe                                                                                                                           |
| `text`     | *string*                                         | :heavy\_check\_mark: | The transcribed text                                                                   | Hello, this is a test of OpenAI speech-to-text transcription. The weather is sunny today and the temperature is around 72 degrees.   |
| `usage`    | [models.STTUsage](../models/sttusage.mdx)        | :heavy\_minus\_sign: | Aggregated usage statistics for the request                                            | \{<br />"cost": 0.000508,<br />"input\_tokens": 83,<br />"output\_tokens": 30,<br />"seconds": 9.2,<br />"total\_tokens": 113<br />} |
| `words`    | [models.STTWord](../models/sttword.mdx)\[]       | :heavy\_minus\_sign: | Timestamped words, present when the provider returns word-level timestamps             |                                                                                                                                      |