# Real-time Processing

Ultravox v0.5 is an audio-to-text model optimized from the Llama-3 2.1B architecture, focusing on efficient speech transcription tasks.

A lightweight portrait segmentation model in ONNX format, specifically optimized for separating people from backgrounds in landscape images.

A lightweight pose estimation model based on ViT architecture for human keypoint detection

SAM 2 Tiny is the Core ML version of the general-purpose segmentation model for images and videos released by FAIR, optimized for mobile applications

This model is a fine-tuned audio classification model based on DistilHuBERT, specifically designed to distinguish between music and non-music audio.

An object detection model based on YOLOv8, specifically designed for detecting handwritten text content

An example vision model for recognizing vehicle license plates, capable of extracting license plate numbers from images.

Open-source model based on Apache-2.0 license, specific functionalities depend on the actual model

This model is used for audio classification tasks and can classify and recognize audio data.

A voice command recognition model based on AST architecture, optimized for web deployment in ONNX format

Speaker diarization model based on pyannote.audio 2.0, used for automatically detecting and segmenting different speakers in audio

A TensorFlow SavedModel-based white-box cartoonizer model capable of converting real images into cartoon-style images.

A small speech recognition model based on the Whisper architecture, suitable for domain-specific speech-to-text tasks

This is an image-to-text model based on the MIT license, capable of converting image content into textual descriptions.

This is an open-source model based on the Apache-2.0 license, with specific functionalities to be determined by the actual model type

This is an open-source model using the Apache-2.0 license, with specific functionalities and application areas requiring further confirmation

An audio classification model based on the wav2vec 2.0 architecture, designed to distinguish between adult and child voices

A speech recognition model fine-tuned based on facebook/wav2vec2-xls-r-300m, supporting left-right channel processing

Keyword recognition model based on the DistilHuBERT architecture, fine-tuned on the superb dataset with an accuracy of 97.06%