--- title: "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 | | | `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 | | `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 | |