dataset

[mshbaita-jo]

I need to understand the format of the dataset I should use ... and what is it contain exactly ?
I mean these files:
"dataset_name": "ylacombe/english_dialects",
"dataset_config_name": "welsh_female",
"override_speaker_embeddings": true,
"filter_on_speaker_id": 5223,

[ylacombe]

Hello,
dataset_name corresponds to the dataset name in the HuggingFace hub, here it is: https://huggingface.co/datasets/ylacombe/english_dialects
dataset_config_name is the name of the "subdataset" you're using, here the Welsh female configuration.

In this dataset, there is a column with speaker ids, so I used filter_on_speaker_id=5223 to filter on the speaker id 5223 and throw out the rest.

override_speaker_embeddings=True is because you want your speaker embeddings to start from scratch instead of continuing training on the previous speaker embeddings of the model

[mshbaita-jo]

The dataset you are using contains the audios and text in one data frame. But for me I have a folder that contains the audios with mp3 format and another TSV file that contains the names of the audios, the texts, and the speaker_id. How can I handle this dataset? and integrate it with the json configuration file?