AI Model Catalogue

Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. Trained on 680k hours of labelled data, Whisper models demonstrate a strong ability to generalize to many datasets and domains without the need for fine-tuning. This is the English-only version of the Whisper Tiny model which was trained on the task of speech recognition.

Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation.

Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multitasking model that can perform multilingual speech recognition, speech translation, and language identification.

Open-source whisper-medium model from openai — available for download and self-hosting on Hugging Face.

Open-source whisper-tiny model from openai — available for download and self-hosting on Hugging Face.

Open-source whisper-large-v3 model from openai — available for download and self-hosting on Hugging Face.

Open-source whisper-large-v3-turbo model from openai — available for download and self-hosting on Hugging Face.

Open-source whisper-base model from openai — available for download and self-hosting on Hugging Face.

Open-source whisper-small model from openai — available for download and self-hosting on Hugging Face.

This lightweight, multimodal model excels at processing text, images, and audio inputs with fast response times and lower costs, making it ideal for developers and organizations building real-time conversational applications, content analysis tools, and interactive voice assistants where efficiency and affordability are priorities.

This multimodal model excels at processing and generating both text and audio, making it ideal for developers and applications requiring real-time voice interactions, transcription, audio analysis, and spoken responses without separate API calls.

This lightweight multimodal model excels at fast, cost-effective processing of text, images, and audio inputs, making it ideal for developers and organizations seeking to build real-time conversational AI applications and voice-enabled features with minimal latency and resource requirements.

I don't have information about a model called "gpt-audio-2025-08-28" in my training data, as my knowledge was last updated in April 2024 and this appears to be a future or non-existent model designation.

This lightweight audio-processing model excels at real-time speech recognition, transcription, and audio understanding tasks, making it ideal for developers building cost-effective voice applications and accessibility features with minimal latency requirements.

The gpt-4o-audio-preview model adds support for audio inputs as prompts. This enhancement allows the model to detect nuances within audio recordings and add depth to generated user experiences. Audio outputs...

A cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Input is priced at $0.60 per million...

The gpt-audio model is OpenAI's first generally available audio model. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Audio is priced...