Dataset Viewer

The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.

LibriSpeech enhanced voice prompts

The voice-cloning prompts used to evaluate pocket-tts / Kyutai TTS models on the LibriSpeech-PC test-clean cross-sentence protocol.

448 unique prompt utterances from LibriSpeech test-clean, speech-enhanced (denoised, 32 kHz mono FLAC, duration-preserving), laid out as <speaker>/<chapter>/<utterance>.flac — the same tree structure as LibriSpeech itself, so --prompt-root style substitution works directly.

Also included: librispeech_pc_test_clean_cross_sentence.lst, the 1127-row cross-sentence eval list (from the F5-TTS evaluation suite): each row pairs a prompt utterance with a target sentence from a different utterance of the same speaker. Column 1 is the prompt utterance id (joins to the FLACs here), columns 4-6 describe the target.

Usage with pocket-tts

python -m training.eval.librispeech <run_dir> \
  --list librispeech_pc_test_clean_cross_sentence.lst \
  --librispeech-root /path/to/LibriSpeech/test-clean \
  --prompt-root .   # this dataset

License / attribution

Derived from LibriSpeech (CC BY 4.0, Panayotov et al.) and the LibriSpeech-PC cross-sentence list distributed with F5-TTS. Released under CC BY 4.0.

Downloads last month
568