Whisper 5

Groq offers fast AI inference through its hardware and software platform for AI applications. Groq is a hardware and software platform that delivers exceptional compute speed, quality, and energy efficiency for AI inference. Groq provides cloud and on-prem solutions at scale for AI applications, offering high-performance AI models and API access for developers. It aims to provide faster inference at a lower cost than competitors.

OpenTypeless is a free, open-source AI voice typing app for macOS, Windows, and Linux that turns speech into polished text in any desktop app. OpenTypeless is an open-source desktop dictation and AI writing assistant. Press a global hotkey, speak naturally, and insert clean text into email, chat, documents, code editors, and other apps. Beyond speech-to-text, it can polish dictated text, adapt the writing style to the current app, translate or rewrite selected text, and answer one-shot voice questions. Users can choose optional managed cloud access or bring their own speech-to-text and language-model providers, including Whisper-compatible endpoints and OpenAI-compatible models. BYOK credentials are stored locally, using the operating system credential vault where available, and requests go directly to the configured provider. The desktop app supports macOS, Windows, and Linux.

Open-source tool to track and analyze OpenAI API usage and costs. A minimalist open-source tool to help you view the cost proportion of various models in your OpenAI API usage for a specific day. It allows you to track and analyze your daily usage and cost of the OpenAI API. Understand the costs for different models, including ChatGPT, GPT-4, Whisper, and text-embedding models. Display the information by time or using a pie chart. This tool is open-source and will not leak your API key.

TextSynth offers access to various AI models via API and playground. TextSynth provides access to large language, text-to-image, text-to-speech, and speech-to-text models like Mistral, Llama, Stable Diffusion, and Whisper through a REST API and a playground. It can be used for text completion, question answering, classification, chat, translation, image generation, speech generation, and speech-to-text transcription. TextSynth employs custom inference code for faster inference and lower costs on standard GPUs and CPUs.

Open-source dictation and transcription software with local processing OpenWhispr is open-source dictation software that converts speech into formatted text. It runs locally on your device, supports local speech-to-text models and optional cloud transcription, and works across macOS, Windows, Linux, and iOS. Users can dictate across apps, transcribe audio, create meeting notes, customize AI instructions, and choose their preferred model. The software supports more than 100 languages, offline use, and custom vocabulary.