How to OpenAI Whisper with per-sentence and per-word timestamp segmentation using DeepInfra

We use essential cookies to make our site work. With your consent, we may also use non-essential cookies to improve user experience and analyze website traffic…

DeepInfra raises $107M Series B to scale the inference cloud — read the announcement

Published on 2023.04.05 by Yessen Kanapin

Getting started

To use DeepInfra's API, you'll need an API key.

Sign up or log in to your DeepInfra account
Navigate to the Dashboard / API Keys section
Create a new API key if you don't have one already

You'll use this API key in your requests to authenticate with our services.

Running speech recognition

Whisper is a Speech-To-Text model from OpenAI. Given an audio file with voice data it produces human speech recognition text with per sentence timestamps. There are different model sizes (small, base, large, etc.) and variants for English, see more at deepinfra.com. By default, Whisper produces by sentence timestamp segmentation. We also host whisper-timestamped that can provide timestamps for words in the audio. You can use it with our REST API. Here's how to use it:

curl -X POST \
  -F "audio=@/home/user/all-in-01.mp3" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  'https://api.deepinfra.com/v1/inference/openai/whisper-timestamped-medium.en'
copy

To see additional parameters and how to call this model, check out the documentation page for complete API reference and examples.

If you have any question, just reach out to us on our Discord server.

MiniMax-M2.5 API Benchmarks: Latency, Throughput & CostAbout MiniMax-M2.5 MiniMax-M2.5 is a state-of-the-art open-weights large language model released in February 2026. Built on a 230B-parameter Mixture of Experts (MoE) architecture with approximately 10 billion active parameters per forward pass, it features Lightning Attention and supports a context window of up to 205,000 tokens. The model uses extended chain-of-thought reasoning to work through […]

Qwen3.5 397B A17B API Benchmarks: Latency, Throughput & CostAbout Qwen3.5 397B A17B Qwen3.5 397B A17B is Alibaba Cloud’s largest and most capable multimodal foundation model, released in February 2026. It features a hybrid Mixture-of-Experts (MoE) architecture with 397 billion total parameters and 17 billion active parameters per inference pass, utilizing 512 experts with a routing mechanism selecting a subset per token. This sparse […]

DeepSeek V3.2 API Benchmarks: Latency, Throughput & CostAbout DeepSeek V3.2 DeepSeek V3.2 is a state-of-the-art large language model that unifies conversational speed and deep reasoning in a single 685B parameter Mixture of Experts (MoE) architecture with 37B parameters activated per token. It is built around three key technical breakthroughs: DeepSeek V3.2 achieved gold-medal performance in the 2025 International Mathematical Olympiad (IMO) and […]

View all