[Marketplace Model] ASR model of noisy telephone data: Wav2Vec2-Large-Robust fine-tuned on Switchboard
Facebook's Wav2Vec2 is a fine-tuned version of the wav2vec2-large-robust model. It has been pre-trained on:
Libri-Light: open-source audiobooks from the LibriVox project; clean, read-out audio data
CommonVoice: crowd-sourced audio data; read-out text snippets
Switchboard: telephone speech corpus; noisy telephone data
Fisher: conversational telephone speech; noisy telephone data
and subsequently been fine-tuned on 300 hours of
Switchboard: telephone speech corpus; noisy telephone data
When using the model, make sure that your speech input is also sampled at 16Khz.
Check out the original research paper for more information, or try it yourself.
[Marketplace Model] ASR models of less common languages Wav2Vec2-XLSR-53
Facebook's XLSR-Wav2Vec2
The base model was pre-trained on 16kHz sampled speech audio. When using the model, make sure that your speech input is also sampled at 16Khz. Note that you should fine-tune this model on a downstream task, like Automatic Speech Recognition. Check out the original research paper for more information.
[Platform-Object] Model: Added MMDetection_YoloF Template for training Visual-Detector
Create a customized training template from reusable training components at the critical stages of Data Preparation, Data Augmentation, Model Configuration, and Model Evaluation. Learn more about training with this and other model architectures with Clarifai Deep Training.
[Platform-Object] Model: Added MMDetection_RetinaNet Template for training Visual-Detector
Create a customized training template from reusable training components at the critical stages of Data Preparation, Data Augmentation, Model Configuration, and Model Evaluation. Learn more about training with this and other model architectures with Clarifai Deep Training.
