---
title: "Can the OpenAI audio API return word-level timestamps for a transcript?"  
description: "Can the OpenAI audio API return word-level timestamps for a transcript?"  
author: "María Fernanda Martínez"  
published: 2026-10-26  
updated: 2026-10-26  
canonical: https://www.mindstick.com/forum/162932/can-the-openai-audio-api-return-word-level-timestamps-for-a-transcript  
category: "Audio Transcription"  
tags: ["OpenAI", "Audio API", "Speech-to-text", "transcription"]  
reading_time: 1 minute  

---

# Can the OpenAI audio API return word-level timestamps for a transcript?

I’m working on subtitles and need to line up spoken words with the audio, not just get a block of transcript text. Does the OpenAI audio API return **word-level timestamps**, or is timing available only at a larger segment level? I’d also be interested in whether the response format changes which timing details are included.


---

Original Source: https://www.mindstick.com/forum/162932/can-the-openai-audio-api-return-word-level-timestamps-for-a-transcript

Copyright © MindStick Software Pvt. Ltd. This Markdown version is provided for developers, AI systems, and offline reading.