This dataset is a 100-sample public-eval smoke split released as raw audio plus a lightweight JSONL manifest.
It is intended for low-cost pipeline bring-up and local fine-tuning smoke tests. It is not the recommended split for formal experiments.
raw/audio/...: the released MP3 audio filesselected_samples.exported.jsonl: sample manifest with
audio_path, target_text, duration, language, and
prompt_boundary_indexexport_info.json: lightweight release metadataThis repository no longer ships prebuilt Arrow artifacts. Prepare the local fine-tuning dataset on your own machine with the corresponding MAGIC-TTS beta data-preparation scripts.
This dataset is part of the MAGIC-TTS release:
If you use this dataset, please cite the paper above so the dataset page is linked with the corresponding research release.
This dataset is a 100-sample public-eval smoke split released as raw audio plus a lightweight JSONL manifest.
It is intended for low-cost pipeline bring-up and local fine-tuning smoke tests. It is not the recommended split for formal experiments.
raw/audio/...: the released MP3 audio filesselected_samples.exported.jsonl: sample manifest with
audio_path, target_text, duration, language, and
prompt_boundary_indexexport_info.json: lightweight release metadataThis repository no longer ships prebuilt Arrow artifacts. Prepare the local fine-tuning dataset on your own machine with the corresponding MAGIC-TTS beta data-preparation scripts.
This dataset is part of the MAGIC-TTS release:
If you use this dataset, please cite the paper above so the dataset page is linked with the corresponding research release.