The advent of artificial intelligence (AI), has brought about a revolution in various fields, including art. One such innovative application of AI in art using Stable Diffusion AI is the creation of the unique Whisper Frame. This AI-powered device listens to conversations and generates relevant images, thereby converting everyday conversations into a unique form of art.
The Whisper Frame is a fascinating blend of technology and creativity. It uses a Raspberry Pi with a re-speaker four-mic array to listen and transcribe conversations. This Raspberry Pi and Respeaker four-mic array setup is an integral part of the device, enabling it to record conversations in a room. The transcriptions are then sent to the OpenAI Whisper API, which generates a transcript of the conversation.
Generates AI art from conversations using Stable Diffusion
The transcript is not the end product but a stepping stone to the final artwork. The transcript is used to generate an image prompt, which is done using GPT-4. GPT-4, a powerful language model, extracts a topic from the conversation transcript and transforms it into an image prompt. This process is a testament to the capabilities of GPT-4, which can understand the context of a conversation and generate relevant prompts.
Once the image prompt is ready, it is sent to Stable Diffusion, an image generator. Stable Diffusion takes the prompt and generates an image based on it. The image generation process is a testament to the capabilities of Stable Diffusion, which can create unique and relevant images based on the prompts.
Other articles you may find of interest on the subject of AI voice and AI art generation technologies :
Whisper Frame powered by Stable Diffusion
The generated images are then displayed on a Samsung Frame. The Samsung Frame is a digital photo frame that adds a touch of elegance to the Whisper Frame. It not only displays the generated images but also enhances the overall aesthetic appeal of the device.
One of the challenges faced during the development of the Whisper Frame was the issue of transcription during silences. To overcome this, the device uses Pico voices Cobra voice activity detection. This software ensures that the device only listens when there is a human voice present in the audio stream, thereby eliminating unnecessary transcriptions during silences.
In addition to generating images, the Whisper Frame also generates an artist’s name for each image and displays the prompt on a sign. This sign is integrated using Adafruit.io as the MQTT broker. The sign adds an extra layer of intrigue to the device, providing viewers with a glimpse into the inspiration behind each image.
The Whisper Frame also has a public website where friends can see what is being displayed in the living room and has been created using Stable Diffusion created by Stability AI. This feature allows the device to share its creations with a wider audience, extending the reach of its unique form of art.
Despite its innovative concept and advanced technology, the Whisper Frame is not always accurate in its image generation. Sometimes, it draws unrelated images, struggles with drawing maps and letters, or interprets prompts in unexpected ways. For instance, it might include a pen in a drawing when the prompt mentioned ink and pen. However, these quirks add a touch of unpredictability and surprise to the device, making it all the more intriguing.
The Whisper Frame has been enjoyed by guests and has been used to generate a variety of images. Its creator has even built a second one for his office, indicating the device’s potential for widespread use. However, it’s worth noting that the Whisper Frame sometimes generates images that are not safe for work, highlighting the need for further refinement in its image generation capabilities.
The Whisper Frame is a unique blend of technology and art, using AI to convert conversations into art displayed on a digital photo frame. Despite its quirks and challenges, it represents a fascinating application of AI in the field of art, opening up new possibilities for creative expression.
Source : WhisperFrame : Hackaday
Filed Under: DIY Projects, Top News
Latest togetherbe Deals
Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, togetherbe may earn an affiliate commission. Learn about our Disclosure Policy.