Text-to-Speech
Transformers
Safetensors
moss_tts_local
feature-extraction
podcast
multi-speaker
voice-cloning
expressive-speech
custom-code
research
custom_code
Instructions to use laion/Humanness-Voice-Podcast with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use laion/Humanness-Voice-Podcast with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="laion/Humanness-Voice-Podcast", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("laion/Humanness-Voice-Podcast", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download stats/release_validation.json from laion/Humanness-Voice-Podcast: direct link, hf CLI and curl.
- Browser
- Download file 2 kB
-
https://huggingface.co/laion/Humanness-Voice-Podcast/resolve/main/stats/release_validation.json
- Command line
-
hf download hf://laion/Humanness-Voice-Podcast/stats/release_validation.json
-
curl -L -o release_validation.json https://huggingface.co/laion/Humanness-Voice-Podcast/resolve/main/stats/release_validation.json
2 kB
| { | |
| "cpu_contract_tests": { | |
| "passed": true, | |
| "count": 16, | |
| "commands": [ | |
| "python -m unittest discover -s tests -v" | |
| ], | |
| "covers": [ | |
| "literal script-end trimming", | |
| "internal hallucination rejection", | |
| "180-ms quiet-pause compaction without modifying words", | |
| "audible gaps retained", | |
| "constant-background crossfade no gain bump", | |
| "400-ms word-clock assembly", | |
| "no double ASR offsets", | |
| "identity-collapse contrast", | |
| "timing/routing format", | |
| "heterogeneous prompt batch CPU fakes", | |
| "CFG1", | |
| "target-only training labels", | |
| "float codec input rejected", | |
| "empty progress journal SQL/render regression", | |
| "populated candidate progress-preview SQL/render regression", | |
| "actual Accelerate sampler partitions across two virtual ranks without an extra tensor dimension" | |
| ] | |
| }, | |
| "source_checkpoint_generation_and_scoring": { | |
| "passed": true, | |
| "candidates": 1536, | |
| "assembled_conditions": 160 | |
| }, | |
| "bf16_export": { | |
| "passed": true, | |
| "proof": "EXPORT_PROOF.json" | |
| }, | |
| "fresh_gpu_smoke": { | |
| "job": 2743115, | |
| "state": "COMPLETED", | |
| "completed": true, | |
| "scope": "Exported BF16 weights + new portable CLI;EN/DE,turns/pairs,two candidates,identity scoring,ASR and quiet-pause option.No VC;not a perceptual quality validation.", | |
| "passed": true, | |
| "time": 1794079841.1333866 | |
| }, | |
| "portable_full_training_run": { | |
| "completed": false, | |
| "historical_full_training_completed": true | |
| }, | |
| "unvalidated": [ | |
| "perceptual effect of internal pause compaction", | |
| "end-to-end optional per-turn voice conversion", | |
| "fresh cross-platform dependency resolver installation" | |
| ], | |
| "time": 1794077542.7875059, | |
| "previous_gpu_smoke_attempts": [ | |
| { | |
| "job": 2742907, | |
| "state": "FAILED", | |
| "phase": "initial progress page before model generation", | |
| "error": "SQL LIMIT4", | |
| "fixed_by": "pipeline-v2, LIMIT 4 plus two regression tests" | |
| } | |
| ] | |
| } |