> ## 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.

# STTSegment

A timestamped transcript segment, returned when response\_format is verbose\_json

## Fields

| Field | Type | Required | Description | Example |
| - | - | - | - | - |
| `avg_logprob` | *Optional\[float]* | :heavy\_minus\_sign: | Average log probability of the segment | |
| `channel` | *Optional\[int]* | :heavy\_minus\_sign: | Zero-based audio channel index for the segment, present when the provider transcribes channels separately | 0 |
| `compression_ratio` | *Optional\[float]* | :heavy\_minus\_sign: | Compression ratio of the segment | |
| `end` | *float* | :heavy\_check\_mark: | Segment end time in seconds | 3.2 |
| `id` | *int* | :heavy\_check\_mark: | Segment index within the transcript | 0 |
| `no_speech_prob` | *Optional\[float]* | :heavy\_minus\_sign: | Probability the segment contains no speech | |
| `seek` | *Optional\[int]* | :heavy\_minus\_sign: | Seek offset of the segment | 0 |
| `speaker` | *Optional\[int]* | :heavy\_minus\_sign: | Speaker index for the segment, present when the provider returns diarization data | 0 |
| `speaker_label` | *Optional\[str]* | :heavy\_minus\_sign: | Provider speaker label for the segment, present when the provider labels speakers with a string | speaker\_0 |
| `start` | *float* | :heavy\_check\_mark: | Segment start time in seconds | 0 |
| `temperature` | *Optional\[float]* | :heavy\_minus\_sign: | Temperature used for the segment | |
| `text` | *str* | :heavy\_check\_mark: | Transcribed text of the segment | Hello there. |
| `tokens` | List\[*int*] | :heavy\_minus\_sign: | Token IDs of the segment | |