diff --git a/README.md b/README.md index 3309060..d313945 100644 --- a/README.md +++ b/README.md @@ -1,10 +1,13 @@ ## Model Description -The model is fine-tuned on the [WavLM base plus](https://arxiv.org/abs/2110.13900) with 2,374 hours of audio clips from +We present a large classification model trained on a manually curated real-world dataset that can be used as a new +benchmark for advancing research in toxicity detection and classification. +Our model is fine-tuned on the [WavLM base plus](https://arxiv.org/abs/2110.13900) with 2,374 hours of audio clips from voice chat for multilabel classification. The audio clips are automatically labeled using a synthetic data pipeline -described in [our blog post](link to blog post here). A single output can have multiple labels. The model outputs a -n by 6 output tensor where the inferred labels are `Profanity`, `DatingAndSexting`, `Racist`, `Bullying`, `Other`, -`NoViolation`. `Other` consists of policy violation categories with low prevalence such as drugs and alcohol or -self-harm that are combined into a single category. +described in [our blog post](link to blog post here). A single output can have multiple labels. +The model outputs a n by 6 output tensor where the inferred labels are `Profanity`, `DatingAndSexting`, `Racist`, +`Bullying`, `Other`, `NoViolation`. `Other` consists of policy violation categories with low prevalence such as drugs +and alcohol or self-harm that are combined into a single category. + We evaluated this model on a data set with human annotated labels that contained a total of 9,795 samples with the class distribution shown below. Note that we did not include the "other" category in this evaluation data set. @@ -18,7 +21,9 @@ distribution shown below. Note that we did not include the "other" category in t |NoViolation | 4185 | 9.93 | 42.73% | -If we set the same threshold across all classes and treat the model as a binary classifier across all 4 toxicity classes (`Profanity`, `DatingAndSexting`, `Racist`, `Bullying`), we get a binarized average precision of 94.48%. The precision recall curve is as shown below. +If we set the same threshold across all classes and treat the model as a binary classifier across all 4 toxicity classes +(`Profanity`, `DatingAndSexting`, `Racist`, `Bullying`), we get a binarized average precision of 94.48%. The precision +recall curve is as shown below.

@@ -35,5 +40,6 @@ To run the inference file, please run the following command: ``` python inference.py --audio_file --model_path ``` -You can get the model weights either by downloading from the model releases page [here](https://github.com/Roblox/voice-safety-classifier/releases/tag/vs-classifier-v1), or from HuggingFace under `roblox/voice-safety-classifier`. -If `model_path` isn’t specified, the model will be loaded directly from HuggingFace. +You can get the model weights either by downloading from the model releases page [here](https://github.com/Roblox/voice-safety-classifier/releases/tag/vs-classifier-v1), or from HuggingFace under +[`roblox/voice-safety-classifier`](https://huggingface.co/Roblox/voice-safety-classifier). If `model_path` isn’t +specified, the model will be loaded directly from HuggingFace.