-
方法名:startRecorder
参数:function(uri?:string,audioSets?: AudioSet,meteringEnabled?: boolean): Promise< string >
不支持属性:uri支持 fd://file.fd, audioSets 支持 Bitrate, Channels, SampleRate, meteringEnabled 不支持
-
# Task Name
Audio Tagging on AudioSet
## Task Objective
This task aims to give an audio some tags that best describe the audio. The model shall give several words (instead of a sentence in au…
-
Firstly I would like to state that this repo is great, so many models, all in pytorch and getting them to work on my machine was very easy.
Have you tried fine-tuning the models on the temporally-s…
-
AudioSet only contains audio and event labels. How do you obtain the caption description for audios in the audioset dataset?
-
AudioSet consists of an expanding ontology of 632 audio event classes and a collection of 2,084,320 human-labeled 10-second sound clips drawn from YouTube videos. The ontology is specified as a hierar…
-
Thank you for the code and inference script.
I understand that the PaSST model has been trained on AudioSet with sampling rate of 32kHz.
I am trying to make inference using the pre trained model.
…
-
-
Dear author,
since the model depends on pretraining on AudioSet to reach the highest score, why not to share the dataset and pretrained-model file? For the audioset dataset always become partiall…
-
* Name of dataset: Google research AudioSet
* URL of dataset: https://research.google.com/audioset/index.html
* License of dataset: Creative Commons
* Short description of dataset and use case(s): …
-