Timestamps

Overview

Soniox Speech-to-Text AI provides precise timestamps for each recognized token (word or sub-word) in your transcription. These timestamps allow you to align text with the original audio for use cases like subtitles, audio indexing, keyword search, and real-time captioning.

Timestamps are returned by default — no configuration is required — and are supported in both asynchronous and real-time processing.

Output format

Each token in the response includes:

• text: The recognized word or token
• start_ms: The start time of the token in milliseconds
• end_ms: The end time of the token in milliseconds

Example response

In this example, the word "beautiful" is split into three tokens with corresponding timestamp ranges.

{
  "tokens": [
    {"text": "Beau", "start_ms": 300, "end_ms": 420},
    {"text": "ti",  "start_ms": 420, "end_ms": 540},
    {"text": "ful", "start_ms": 540, "end_ms": 780}
  ]
}

Use cases