This free & open source extension enables voice interaction with ChatGPT and Clude.ai () in Chrome browser and other Chromium-based browsers (e.g. Edge). It allows the use of the AI model "Whisper" developed by OpenAI for voice to text transcription. You can also use the transcription method built into the browser (webkitSpeechRecognition). The extension appears and operates upon visiting chat.openai.com or clude.ai

Program features

• you can talk to the chat by speaking into the microphone and its responses will be read out loud
• you can enable an option that will read your entire chat conversation out loud, or only read the last response aloud
• you can highlight a text fragment in the chat thread and enable a feature to read this fragment out loud
• in the extension configuration, you can set voice parameters, including the language in which you converse with the chat, voice, voice pitch and reading speed, voice transcription method

STT Voice Transcription (speech-to-text)

The extension offers two methods of voice transcription:

1. webkitSpeechRecognition - the default method using Chrome browser's API. Transcription is performed locally (offline). Note: This method does not add punctuation and is supported only in Chrome browser.

2. Whisper - an AI model developed by OpenAI for speech to text transcription. It offers high-quality transcription with proper punctuation. Requires an OpenAI API key, which involves costs (OpenAI pricing). Transcription is performed on OpenAI servers, which requires sending the audio recording.

TTS Speech Synthesis (text-to-speech)