Text-to-Speech
AudioSeal
Safetensors
English
zero-shot
voice-cloning
english
flow-matching
diffusion-transformer
dacvae
Instructions to use VoiceHub/DACFlow-EN-10k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- AudioSeal
How to use VoiceHub/DACFlow-EN-10k with AudioSeal:
# Watermark Generator from audioseal import AudioSeal model = AudioSeal.load_generator("VoiceHub/DACFlow-EN-10k") # pass a tensor (tensor_wav) of shape (batch, channels, samples) and a sample rate wav, sr = tensor_wav, 16000 watermark = model.get_watermark(wav, sr) watermarked_audio = wav + watermark# Watermark Detector from audioseal import AudioSeal detector = AudioSeal.load_detector("VoiceHub/DACFlow-EN-10k") result, message = detector.detect_watermark(watermarked_audio, sr) - Notebooks
- Google Colab
- Kaggle
Download showcase.json from VoiceHub/DACFlow-EN-10k: direct link, hf CLI and curl.
- Browser
- Download file 49.2 kB
-
https://huggingface.co/VoiceHub/DACFlow-EN-10k/resolve/main/showcase.json
- Command line
-
hf download hf://VoiceHub/DACFlow-EN-10k/showcase.json
-
curl -L -o showcase.json https://huggingface.co/VoiceHub/DACFlow-EN-10k/resolve/main/showcase.json
49.2 kB
| { | |
| "card_sha": "f2ad43054ed7dfca", | |
| "checkpoints": { | |
| "en_base/step_0020000": { | |
| "added": "2026-10-03T02:01:41Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "965a922f22b436056f44641310dbc7d8507a2c899327957c24fe231481c1f16d" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "c54c4224f781d586b616eb1585a97cb0609042e2f9b3b7d57ac0d9a3ab3717ee" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0020000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7218720472775972, | |
| "utmos": 3.33140557545882, | |
| "wer": 0.011970534069981584 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.505644196367592, | |
| "utmos": 3.2711969318739866, | |
| "wer": 0.024732620320855617 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00020000.pt", | |
| "step": 20000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0030000": { | |
| "added": "2026-10-03T02:01:09Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "6ac576a9ea1224a9cbf1b7aed163625cb2b3b9260e4e1977ef4cf8eabd510f3e" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "c7d9769fec397d3011530cbe1acc0ce532693bc2733db9b66bf4516f769c044d" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0030000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7406449794769288, | |
| "utmos": 3.4309559143506565, | |
| "wer": 0.010589318600368325 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00030000.pt", | |
| "step": 30000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0040000": { | |
| "added": "2026-10-03T02:00:35Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "d2532d133fa19ddeb473b1ee850d45aff83242272d9a4439e9a7295e71feab6a" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "8b0c4db624506acceef5669ffb45a604cd04c3c27ac8ac486e35539aa65d690a" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0040000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7480492035929973, | |
| "utmos": 3.5002765545478236, | |
| "wer": 0.007366482504604052 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.5361745800857151, | |
| "utmos": 3.476590223049899, | |
| "wer": 0.018549465240641713 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00040000.pt", | |
| "step": 40000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0050000": { | |
| "added": "2026-10-03T03:02:28Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "0ab3e8e12f06d2637eaf8092d9ab5779571abd19af11df891a0a706de257c2ae" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "d77ce09b7f703b333fa5ac5d6cee67dd6bfb2c5ae6619fbb091845408bd8e8fd" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0050000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.754027674977596, | |
| "utmos": 3.5555466156739457, | |
| "wer": 0.007366482504604052 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00050000.pt", | |
| "step": 50000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0060000": { | |
| "added": "2026-10-03T05:43:27Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "9b6dd095f7543088bc65742ac5bc5a2d20a881da7fe45d326d380a85cd41b1a6" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "f6283084426bb6f3af8c1428657c6932b844bdba04eda6829c74cee3690214df" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0060000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7593376639943856, | |
| "utmos": 3.6220227929262014, | |
| "wer": 0.006906077348066298 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.5368829132698545, | |
| "utmos": 3.561331022332568, | |
| "wer": 0.01612633689839572 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00060000.pt", | |
| "step": 60000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0070000": { | |
| "added": "2026-10-03T07:04:18Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "80d5d578162a77f57e6e8b90ad403471460f50bf83b27fce4b5f3f6daddd5232" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "06abc5dac254b8f89bcee97e5d694b54e4a5ee8a4b1dfcada4a1987ca29764df" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0070000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7662215784192086, | |
| "utmos": 3.688956035100497, | |
| "wer": 0.006675874769797422 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00070000.pt", | |
| "step": 70000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0080000": { | |
| "added": "2026-10-03T09:05:19Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "955d0e3a53bfc449408dc015956e384a1092c2e8ff77f40a3f90a6bd8aac7433" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "410ac3cbfaa1eb83b3226c980247262ea95a270bbf034b4db18c5bcad738787f" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0080000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7762147039175034, | |
| "utmos": 3.7316342271291294, | |
| "wer": 0.007826887661141806 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.50722517402236, | |
| "utmos": 3.6791523234559858, | |
| "wer": 0.01579211229946524 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00080000.pt", | |
| "step": 80000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0090000": { | |
| "added": "2026-10-03T11:06:19Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5713, | |
| "sha256": "457315876e47886708c4f8620d739ec5b7caedcaad69410a3f7ac9988bfa9c90" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "e273799529241fea517cd703b41bf15f21fb86237fa955c868201adc1e88f201" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0090000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7816013838236149, | |
| "utmos": 3.7776074923001803, | |
| "wer": 0.007366482504604052 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00090000.pt", | |
| "step": 90000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0100000": { | |
| "added": "2026-10-03T13:07:27Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "5c3f07cfabce41887989cc6acb726251b584f6cc38a333c31c6655c751c81034" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "9fe92a1f3c3abd5c845979a147cc38cb350c09be0ef93abcdc97e78b6b610efe" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0100000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7872023932062663, | |
| "utmos": 3.851715524380024, | |
| "wer": 0.006906077348066298 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.4993639859926263, | |
| "utmos": 3.7841195987998892, | |
| "wer": 0.01537433155080214 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00100000.pt", | |
| "step": 100000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0110000": { | |
| "added": "2026-10-03T15:48:57Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "d152712a462a03628c83088df00a2ef5130090de257896a892f83ef389c83172" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "4bb806fafa14eba16dfc249add6f6d79d2d5438a890fea6904533c40554f842c" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0110000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.78920833835235, | |
| "utmos": 3.8616720410493706, | |
| "wer": 0.011049723756906077 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00110000.pt", | |
| "step": 110000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0120000": { | |
| "added": "2026-10-03T17:30:04Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "c404fa45e6dc253323a7a4c2a16587ce12579f619607b29cf64bc9cf5bd4593a" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "a7c2759b043ba1a2d26d30bbaeb53dd9e992e4fe055cc7ae10cdf126a44dcea2" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0120000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.790249391244008, | |
| "utmos": 3.8890885912455047, | |
| "wer": 0.005294659300184162 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.4509710253147972, | |
| "utmos": 3.860558075314268, | |
| "wer": 0.014455213903743316 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00120000.pt", | |
| "step": 120000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0130000": { | |
| "added": "2026-10-03T19:31:41Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "9580ef1c5032468582bd16d5f390ecfd1a67dfdbb508ce401e02142da9d72580" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "5489a78587219a357433cc145f3e236a26f19d7ac48c597b8ff94ba596a0d3e6" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0130000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7915309699682089, | |
| "utmos": 3.9265544396180374, | |
| "wer": 0.007596685082872928 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00130000.pt", | |
| "step": 130000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0140000": { | |
| "added": "2026-10-03T21:33:08Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "a82191b3c9a774e7aec3a108f5871e36eeb53c17584dc98a1a829bf7e7d0fb99" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "0a652fd0d411fd9fb65fd0d03028ab7076c0ddbdaed29fcd75ed2bfb1c815b9a" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0140000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7928165781956452, | |
| "utmos": 3.917515279696538, | |
| "wer": 0.005755064456721915 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.4041899165728155, | |
| "utmos": 3.8050217373655477, | |
| "wer": 0.015959224598930483 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00140000.pt", | |
| "step": 140000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0150000": { | |
| "added": "2026-10-04T00:14:14Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "a37b6985556773fc548ff7697516c084046732c58bb8e05a7f99da621ed5f644" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "e30d138d30eda5517f6fa45d2a60c9d18024c844741299854c0adc66c9fa3a44" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0150000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7937824732982195, | |
| "utmos": 3.927360293498406, | |
| "wer": 0.005294659300184162 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00150000.pt", | |
| "step": 150000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0160000": { | |
| "added": "2026-10-04T01:55:12Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "49ba4127427d513388c6bb889ba6587e8efd2ce39fb420db1aa0db1bfedf0d7a" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "ba1a5bdf78e23a7e784c325dc396ff9a5f6f221252fadf8c990f8baa3eb54475" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0160000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7963264559323971, | |
| "utmos": 3.9782051058915946, | |
| "wer": 0.008057090239410682 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.3489067445312618, | |
| "utmos": 3.6552217593980494, | |
| "wer": 0.01579211229946524 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00160000.pt", | |
| "step": 160000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0170000": { | |
| "added": "2026-10-04T03:56:10Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "317d56588e3dd93ecb48a9a53be8b50b01403909f5f6441530d22b9f7cbfe4d9" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "25c3d47168eb03b6ab375143f7c9d19687e50590334d5c9b4a2401123c98b0ec" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0170000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7958178685261653, | |
| "utmos": 3.9733734222558827, | |
| "wer": 0.010128913443830571 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00170000.pt", | |
| "step": 170000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0180000": { | |
| "added": "2026-10-04T05:57:11Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "6c405b3fda26a14e5b83bc8095fbe2b29b13e746629104bd8b4224650178fdf0" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "952034355a488fcc440092052f1bad99e7358d1ea497a22c292c8743e7ec6e37" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0180000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7919172683587441, | |
| "utmos": 3.9800812565363373, | |
| "wer": 0.010359116022099447 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sim_o": 0.26134766056586844, | |
| "utmos": 3.2862812184412546, | |
| "wer": 0.016460561497326203 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00180000.pt", | |
| "step": 180000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| }, | |
| "en_base/step_0190000": { | |
| "added": "2026-10-04T08:18:19Z", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "f0909e4", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5714, | |
| "sha256": "5e6826eed32a7c1f13e2fa3ee90063e1cb81a14bae241a8e19913dd7520fd1d2" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825673000, | |
| "sha256": "e0c56f0b661b7c8f6ed2a2fe9c46a325511fe7509bfc0c2f186b6a6a6c0f2b6f" | |
| } | |
| }, | |
| "folder": "checkpoints/step_0190000", | |
| "lineage": [], | |
| "metrics": { | |
| "echo_dev": { | |
| "n": 260, | |
| "sim_o": 0.7896125185948152, | |
| "utmos": 3.980060512285966, | |
| "wer": 0.006445672191528545 | |
| } | |
| }, | |
| "note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "recipe_arm": false, | |
| "run": "en_base", | |
| "source": "en_base/export/model_step_00190000.pt", | |
| "step": 190000, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "verdict": "-", | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| } | |
| }, | |
| "decoders": { | |
| "ft_A_u300_step_005000": { | |
| "ab": { | |
| "echo_dev_v2": { | |
| "sim_o": { | |
| "ci": [ | |
| 0.002, | |
| 0.004 | |
| ], | |
| "diff": 0.003, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.049, | |
| 0.062 | |
| ], | |
| "diff": 0.055, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| -0.038, | |
| 0.108 | |
| ], | |
| "diff": 0.03, | |
| "significant": false | |
| } | |
| }, | |
| "seed_dev": { | |
| "sim_o": { | |
| "ci": [ | |
| -0.008, | |
| -0.005 | |
| ], | |
| "diff": -0.006, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.055, | |
| 0.068 | |
| ], | |
| "diff": 0.061, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| -0.137, | |
| -0.016 | |
| ], | |
| "diff": -0.07, | |
| "significant": true | |
| } | |
| }, | |
| "voice_quality": { | |
| "hnr_db": [ | |
| 0.57, | |
| 0.01 | |
| ], | |
| "shimmer": [ | |
| -0.0048, | |
| 0.0002 | |
| ] | |
| } | |
| }, | |
| "added": "2026-10-07T00:41:06Z", | |
| "base_codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "code_commit": "f22a0cc", | |
| "decoder_sha256": "071d06c9488dd48dae02bb4c0745afc96bea7b2177996d17f906d0751b72b115", | |
| "encoder_sha256": "e850278bc478910d809ff3986b56324b2cfb29294e5cbfebd97027530935a9b2", | |
| "files": { | |
| "README.md": { | |
| "bytes": 1617, | |
| "sha256": "bb92d5a63bb1c753aaf96d8da5adf47e6b3a7080a3c7b4a2719913b531328e2a" | |
| }, | |
| "config.json": { | |
| "bytes": 2251, | |
| "sha256": "ab8b6698bb1d7d7f64cdf369efde157cd91b8f25a6effb432fb969fc67d3ccca" | |
| }, | |
| "decoder.safetensors": { | |
| "bytes": 320500876, | |
| "sha256": "14675eaaaf878ee2214f5b5c07091e57c08a3c3474d080aa3155d8a5170d88d1" | |
| } | |
| }, | |
| "fitted_to": "models/grow_q_u300", | |
| "folder": "decoders/ft_A_u300_step_005000", | |
| "name": "ft_A_u300_step_005000", | |
| "source_codec_sha": "9ad732326b3e6b90", | |
| "step": 5000, | |
| "title": "ft_A, 5k updates" | |
| }, | |
| "ft_A_u300_step_010000": { | |
| "ab": { | |
| "echo_dev_v2": { | |
| "sim_o": { | |
| "ci": [ | |
| 0.003, | |
| 0.005 | |
| ], | |
| "diff": 0.004, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.059, | |
| 0.072 | |
| ], | |
| "diff": 0.065, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| -0.01, | |
| 0.179 | |
| ], | |
| "diff": 0.08, | |
| "significant": false | |
| } | |
| }, | |
| "seed_dev": { | |
| "sim_o": { | |
| "ci": [ | |
| -0.011, | |
| -0.008 | |
| ], | |
| "diff": -0.01, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.058, | |
| 0.072 | |
| ], | |
| "diff": 0.065, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| -0.137, | |
| 0.034 | |
| ], | |
| "diff": -0.05, | |
| "significant": false | |
| } | |
| }, | |
| "voice_quality": { | |
| "hnr_db": [ | |
| 0.65, | |
| 0.02 | |
| ], | |
| "shimmer": [ | |
| -0.0057, | |
| 0.0002 | |
| ] | |
| } | |
| }, | |
| "added": "2026-10-07T00:41:07Z", | |
| "base_codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "code_commit": "f22a0cc", | |
| "decoder_sha256": "4a5786a8be6a0b786a5038a61fa3816a5040b1b4fb422ca255099694115a12ac", | |
| "encoder_sha256": "e850278bc478910d809ff3986b56324b2cfb29294e5cbfebd97027530935a9b2", | |
| "files": { | |
| "README.md": { | |
| "bytes": 1625, | |
| "sha256": "16ae8d11a9b9c8fc6bec106939b4958ac35efa4476d70bd3d467f949ca712843" | |
| }, | |
| "config.json": { | |
| "bytes": 2254, | |
| "sha256": "4332242bbff1a92a6c238b83b92fcaf318361dc0730da7173537398f016e3d68" | |
| }, | |
| "decoder.safetensors": { | |
| "bytes": 320500876, | |
| "sha256": "2df57b85e682da901d9969f40d3cf545662325f50ae589e530f83e58143af8b8" | |
| } | |
| }, | |
| "fitted_to": "models/grow_q_u300", | |
| "folder": "decoders/ft_A_u300_step_010000", | |
| "name": "ft_A_u300_step_010000", | |
| "source_codec_sha": "c8407490a1f63b41", | |
| "step": 10000, | |
| "title": "ft_A, 10k updates" | |
| }, | |
| "ft_A_u300_step_020000": { | |
| "ab": { | |
| "echo_dev_v2": { | |
| "sim_o": { | |
| "ci": [ | |
| 0.005, | |
| 0.007 | |
| ], | |
| "diff": 0.006, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.07, | |
| 0.083 | |
| ], | |
| "diff": 0.077, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| 0.0, | |
| 0.118 | |
| ], | |
| "diff": 0.05, | |
| "significant": false | |
| } | |
| }, | |
| "seed_dev": { | |
| "sim_o": { | |
| "ci": [ | |
| -0.011, | |
| -0.008 | |
| ], | |
| "diff": -0.01, | |
| "significant": true | |
| }, | |
| "utmos": { | |
| "ci": [ | |
| 0.072, | |
| 0.086 | |
| ], | |
| "diff": 0.079, | |
| "significant": true | |
| }, | |
| "wer": { | |
| "ci": [ | |
| -0.13, | |
| 0.017 | |
| ], | |
| "diff": -0.05, | |
| "significant": false | |
| } | |
| }, | |
| "voice_quality": { | |
| "hnr_db": [ | |
| 0.76, | |
| 0.02 | |
| ], | |
| "shimmer": [ | |
| -0.0068, | |
| 0.0002 | |
| ] | |
| } | |
| }, | |
| "added": "2026-10-07T00:41:08Z", | |
| "base_codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "code_commit": "f22a0cc", | |
| "decoder_sha256": "60aa04ac8660d5647f8bac2a21601c065a2b7c9aa910a0dc2e4aeef066c47fb8", | |
| "encoder_sha256": "e850278bc478910d809ff3986b56324b2cfb29294e5cbfebd97027530935a9b2", | |
| "files": { | |
| "README.md": { | |
| "bytes": 1625, | |
| "sha256": "f170c73d3912e4ade054211f4dc46182194f9912aec4d152ddb7d3d1bd1b8ed6" | |
| }, | |
| "config.json": { | |
| "bytes": 2254, | |
| "sha256": "f63d22e4ebc1ed85329dd3569d31e168be56233d08245b29192ae74626ca4ab8" | |
| }, | |
| "decoder.safetensors": { | |
| "bytes": 320500876, | |
| "sha256": "1de4b3e078a7edd7700bcf882ce46a62187c669475108f0a24c32c24d22616e0" | |
| } | |
| }, | |
| "fitted_to": "models/grow_q_u300", | |
| "folder": "decoders/ft_A_u300_step_020000", | |
| "name": "ft_A_u300_step_020000", | |
| "source_codec_sha": "632e049ba9713ffe", | |
| "step": 20000, | |
| "title": "ft_A, 20k updates" | |
| } | |
| }, | |
| "decoders_info": { | |
| "ab": { | |
| "base": { | |
| "echo_dev_v2": { | |
| "n": 1000, | |
| "sim_o": 0.82, | |
| "utmos": 4.04, | |
| "wer_pct": 0.47 | |
| }, | |
| "seed_dev": { | |
| "n": 545, | |
| "sim_o": 0.54, | |
| "utmos": 3.855, | |
| "wer_pct": 1.39 | |
| } | |
| }, | |
| "voice_quality": { | |
| "base": { | |
| "hnr_db": 10.5385, | |
| "shimmer": 0.1029 | |
| }, | |
| "items": 1000, | |
| "set": "echo_dev_v2" | |
| } | |
| }, | |
| "base_codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "base_codec_licence": "MIT", | |
| "default": "ft_A_u300_step_020000", | |
| "fitted_to": "grow_q_u300", | |
| "pull": 184, | |
| "training": { | |
| "catalog": "echo_en_v1", | |
| "catalog_version": "data-4255c6734a7d5b5e", | |
| "dataset": "SynDataLab-EN/echo-clones-4m-en", | |
| "excluded_voices": 413, | |
| "hours": 39.29, | |
| "modes": [ | |
| "sdedit@0.5", | |
| "sdedit@0.6" | |
| ], | |
| "pairs": 20000, | |
| "voices": 3581 | |
| } | |
| }, | |
| "featured": { | |
| "bandwidth": "auto", | |
| "bandwidth_range_from": "p5..p95 of the training catalog echo_en_v1 (data-4255c6734a7d5b5e, 3,397,565 train clips), as evaluate.py --bandwidth auto recorded it", | |
| "bandwidth_range_hz": [ | |
| 9797.607421875, | |
| 16838.96484375 | |
| ], | |
| "base_key": "en_base/step_0100000", | |
| "compare": { | |
| "bandwidth": "default", | |
| "checkpoints": [ | |
| { | |
| "echo_dev_v2": { | |
| "n": 1000, | |
| "share_sim_o_below_0_2": 0.0, | |
| "sim_o": 0.816516, | |
| "utmos": 4.00165, | |
| "wer": 0.004919 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "share_sim_o_below_0_2": 0.0569, | |
| "sim_o": 0.499364, | |
| "utmos": 3.78412, | |
| "wer": 0.015374 | |
| }, | |
| "step": 100000 | |
| }, | |
| { | |
| "echo_dev_v2": { | |
| "n": 1000, | |
| "share_sim_o_below_0_2": 0.001, | |
| "sim_o": 0.820022, | |
| "utmos": 4.137699, | |
| "wer": 0.004575 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "share_sim_o_below_0_2": 0.5147, | |
| "sim_o": 0.222872, | |
| "utmos": 3.004047, | |
| "wer": 0.020555 | |
| }, | |
| "step": 190000 | |
| } | |
| ] | |
| }, | |
| "decided": "2026-10-04: owner decision, the M5 base is the step-100k EMA export with --bandwidth auto", | |
| "duration_mode": "band", | |
| "file": "en_release_featured.json", | |
| "issue": 177, | |
| "key": "models/grow_q_u300", | |
| "model": { | |
| "adopted": "It passed a rule fixed before its evaluation: higher UTMOS and DNSMOS on echo-dev v2 (95 % intervals above 0), word error rate not worse on either set, seed-dev SIM-o lower by at most 0.01.", | |
| "base_scores": { | |
| "echo_dev_v2": { | |
| "n": 1000, | |
| "sentences": 1000, | |
| "share_sim_o_below_0_2": 0.0, | |
| "sim_o": 0.816373, | |
| "takes": 1, | |
| "utmos": 3.9925, | |
| "voices": 198, | |
| "wer": 0.005264 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sentences": 545, | |
| "share_sim_o_below_0_2": 0.0239, | |
| "sim_o": 0.526291, | |
| "takes": 2, | |
| "utmos": 3.804776, | |
| "wer": 0.014205 | |
| } | |
| }, | |
| "decided": "2026-10-06: the EN-45 post-round verdict ADOPT under the pre-registered rule EN-45-quality-adopt-v2; the owner approved featuring it on the page", | |
| "export_sha256": "6f1659f17cb6b86f4206c41741ee80f2f5cab3bdf72a3f55b7a0cc32426bbc59", | |
| "export_step": 300, | |
| "issue": 69, | |
| "name": "grow_q_u300", | |
| "recipe": "GROW, an online reinforcement-learning method for flow-matching models: for each of 16 prompts per update the model reads the text 16 times, each reading is scored (intelligibility by a CTC speech recognizer, voice similarity by a speaker-verification model, predicted naturalness by UTMOS), and the model is moved toward its better readings, with a penalty that keeps it close to the base (learning rate 2e-6). The prompts are 25,000 texts read by 320 voices of the training data (its transcripts and voices plus templated hard sentences with numbers, names and abbreviations): no new audio, only the model's own readings and the scores of pretrained scoring models. The round was planned for 500 updates and stopped at update 304 when the machine restarted; of the saved updates (100, 200, 300), the one with the lowest word error rate on a held-out dev list, update 300, was kept, as the rule fixed before the round said.", | |
| "title": "100k base + EN-45 quality post-training (300 updates)" | |
| }, | |
| "run": "en_base", | |
| "scores": { | |
| "echo_dev_v2": { | |
| "n": 1000, | |
| "sentences": 1000, | |
| "share_sim_o_below_0_2": 0.0, | |
| "sim_o": 0.81989, | |
| "takes": 1, | |
| "utmos": 4.040244, | |
| "voices": 198, | |
| "wer": 0.005018 | |
| }, | |
| "seed_dev": { | |
| "n": 1090, | |
| "sentences": 545, | |
| "share_sim_o_below_0_2": 0.0174, | |
| "sim_o": 0.536487, | |
| "takes": 2, | |
| "utmos": 3.86231, | |
| "wer": 0.014205 | |
| } | |
| }, | |
| "step": 100000, | |
| "stopped_at": 194271, | |
| "stopped_on": "2026-10-04" | |
| }, | |
| "format": "mytts-showcase/5", | |
| "items": [ | |
| { | |
| "category": "short", | |
| "item": "01-short", | |
| "pitch": "higher", | |
| "prompt_clip": "train-00361-486", | |
| "prompt_f0_hz": 201.8, | |
| "prompt_license": "apache-2.0 (SynDataLab-EN/echo-clones-4m-en, held-out dev voice)", | |
| "prompt_seconds": 9.242, | |
| "prompt_sha256": "5d1822e59071fe22d6910d08f4c2c49735853be3cf50916b1e7cef17c246f477", | |
| "prompt_text": "I keep telling myself I just need to power through, like, push a little harder, and then it'll be fine. But it's not getting fine. It's getting— it's getting worse.", | |
| "seed": 0, | |
| "text": "I left my umbrella at the office again, so I'm definitely getting soaked on the way home.", | |
| "voice": "spk_0610", | |
| "voice_set": "echo" | |
| }, | |
| { | |
| "category": "question", | |
| "item": "02-question", | |
| "pitch": "lower", | |
| "prompt_clip": "train-00024-33", | |
| "prompt_f0_hz": 119.3, | |
| "prompt_license": "apache-2.0 (SynDataLab-EN/echo-clones-4m-en, held-out dev voice)", | |
| "prompt_seconds": 5.712, | |
| "prompt_sha256": "ec5010cbb72864e62cd95a87840c6f74a750f4ac135ddcda43ce34417b63c346", | |
| "prompt_text": "Look, I'm not saying it's easy, but you always pull it together. Just take it step by step, you know?", | |
| "seed": 0, | |
| "text": "Have you ever noticed that the quietest person in the room usually has the most interesting story to tell?", | |
| "voice": "spk_0327", | |
| "voice_set": "echo" | |
| }, | |
| { | |
| "category": "numbers", | |
| "item": "03-numbers", | |
| "pitch": "higher", | |
| "prompt_clip": "train-02023-409", | |
| "prompt_f0_hz": 200.7, | |
| "prompt_license": "apache-2.0 (SynDataLab-EN/echo-clones-4m-en, held-out dev voice)", | |
| "prompt_seconds": 7.198, | |
| "prompt_sha256": "1d667de2ea2cc8713b440ee979664f94031b4148333a5a61d59de5946d0de008", | |
| "prompt_text": "Would you just— I know you mean well, but every time you mention it I feel like an idiot. It's probably nothing.", | |
| "seed": 0, | |
| "text": "Dr. Patel moved my appointment to Tuesday, March 3rd, at 4:15 p.m., and the co-pay went up from $20 to $35.", | |
| "voice": "spk_0409", | |
| "voice_set": "echo" | |
| }, | |
| { | |
| "category": "conversational", | |
| "item": "04-conversational", | |
| "pitch": "lower", | |
| "prompt_clip": "train-02182-257", | |
| "prompt_f0_hz": 114.6, | |
| "prompt_license": "apache-2.0 (SynDataLab-EN/echo-clones-4m-en, held-out dev voice)", | |
| "prompt_seconds": 4.272, | |
| "prompt_sha256": "b87421add444ab00359a0cf5eb3b6bd81dd6a60156b5ffbffe07a80b1af9f4c0", | |
| "prompt_text": "I'm sorry, but the subway doors closed on my bag. Again.", | |
| "seed": 0, | |
| "text": "So I finally tried that new ramen place downtown, and honestly, it was worth the wait. The broth was amazing, but next time I'm definitely skipping the extra spicy option.", | |
| "voice": "spk_3757", | |
| "voice_set": "echo" | |
| }, | |
| { | |
| "category": "long", | |
| "item": "05-long", | |
| "pitch": "lowest", | |
| "prompt_clip": "train-09627-464", | |
| "prompt_f0_hz": 102.1, | |
| "prompt_license": "apache-2.0 (SynDataLab-EN/echo-clones-4m-en, held-out dev voice)", | |
| "prompt_seconds": 8.034, | |
| "prompt_sha256": "8311aeaf5ac99a1b940d1945c3390de11f845c02328163fe4fadad8e7943c4e0", | |
| "prompt_text": "I mean, come on, it's not like I woke up and decided to have the worst day ever. Things just went wrong one after another, like dominoes!", | |
| "seed": 0, | |
| "text": "When the storm passed, the whole neighborhood came outside to look at the damage, and although a few fences had fallen and the old oak tree had lost its biggest branch, everyone was relieved that nobody was hurt. They spent the rest of the afternoon clearing the street and sharing whatever food they had left.", | |
| "voice": "spk_1964", | |
| "voice_set": "echo" | |
| } | |
| ], | |
| "items_sha": "8fcd29112a5e183c", | |
| "kind": "release", | |
| "models": { | |
| "grow_q_u300": { | |
| "added": "2026-10-06T17:02:10Z", | |
| "base": "en_base/step_0100000", | |
| "catalog": "echo_en_v1", | |
| "catalog_clips": 3400911, | |
| "catalog_hours": 9543.28, | |
| "catalog_voices": 3587, | |
| "code_commit": "7af9fac", | |
| "codec_id": "Aratako/Semantic-DACVAE-Japanese@737fbad2a679f880", | |
| "data_version": "data-4255c6734a7d5b5e", | |
| "export_sha256": "6f1659f17cb6b86f4206c41741ee80f2f5cab3bdf72a3f55b7a0cc32426bbc59", | |
| "files": { | |
| "config.json": { | |
| "bytes": 5711, | |
| "sha256": "ab393d7ceffee04b709d637648e94c59cae0eeab5727cc9f2e93581640080f75" | |
| }, | |
| "model.safetensors": { | |
| "bytes": 825672992, | |
| "sha256": "fd00076e87ec51fd1ce1b29140c383052db07aed7ae4be25e50781f3030de96c" | |
| } | |
| }, | |
| "folder": "checkpoints/grow_q_u300", | |
| "lineage": [], | |
| "model": "grow_q_u300", | |
| "params_m": 206.41, | |
| "preset": "base", | |
| "provenance": "catalog echo_en_v1 (data-4255c6734a7d5b5e): echo_en", | |
| "run": "en_grow_q", | |
| "samples": { | |
| "added": "2026-10-16T20:43:53Z", | |
| "code_commit": "f4042c8", | |
| "folder": "samples/grow_q_u300", | |
| "items_sha": "8fcd29112a5e183c", | |
| "rows": [ | |
| { | |
| "bandwidth_clamped": null, | |
| "bandwidth_hz": 15568.5, | |
| "file": "samples/grow_q_u300/01-short.wav", | |
| "gen_seconds": 1.91, | |
| "item": "01-short", | |
| "prompt_bandwidth_hz": 15568.5, | |
| "seconds": 5.12, | |
| "watermark": "audioseal_wm_16bits", | |
| "watermark_bits": "0100110101010100", | |
| "watermark_score": 0.985 | |
| }, | |
| { | |
| "bandwidth_clamped": "high", | |
| "bandwidth_hz": 16839.0, | |
| "file": "samples/grow_q_u300/02-question.wav", | |
| "gen_seconds": 0.72, | |
| "item": "02-question", | |
| "prompt_bandwidth_hz": 16968.2, | |
| "seconds": 6.08, | |
| "watermark": "audioseal_wm_16bits", | |
| "watermark_bits": "0100110101010100", | |
| "watermark_score": 1.0 | |
| }, | |
| { | |
| "bandwidth_clamped": null, | |
| "bandwidth_hz": 15482.4, | |
| "file": "samples/grow_q_u300/03-numbers.wav", | |
| "gen_seconds": 0.76, | |
| "item": "03-numbers", | |
| "prompt_bandwidth_hz": 15482.4, | |
| "seconds": 8.76, | |
| "watermark": "audioseal_wm_16bits", | |
| "watermark_bits": "0100110101010100", | |
| "watermark_score": 1.0 | |
| }, | |
| { | |
| "bandwidth_clamped": null, | |
| "bandwidth_hz": 16559.0, | |
| "file": "samples/grow_q_u300/04-conversational.wav", | |
| "gen_seconds": 0.78, | |
| "item": "04-conversational", | |
| "prompt_bandwidth_hz": 16559.0, | |
| "seconds": 11.76, | |
| "watermark": "audioseal_wm_16bits", | |
| "watermark_bits": "0100110101010100", | |
| "watermark_score": 1.0 | |
| }, | |
| { | |
| "bandwidth_clamped": null, | |
| "bandwidth_hz": 15697.7, | |
| "file": "samples/grow_q_u300/05-long.wav", | |
| "gen_seconds": 1.54, | |
| "item": "05-long", | |
| "prompt_bandwidth_hz": 15697.7, | |
| "seconds": 17.62, | |
| "watermark": "audioseal_wm_16bits", | |
| "watermark_bits": "0100110101010100", | |
| "watermark_score": 1.0 | |
| } | |
| ], | |
| "sampler": { | |
| "bandwidth": "auto", | |
| "bandwidth_range_hz": [ | |
| 9797.607421875, | |
| 16838.96484375 | |
| ], | |
| "cfg_mode": "joint", | |
| "cfg_rescale": 0.0, | |
| "cfg_t_max": 1.0, | |
| "cfg_t_min": 0.0, | |
| "cfg_w": 4.0, | |
| "gpu_peak_gb": 3.27, | |
| "noise_scale": 0.9, | |
| "out_lufs": -16.0, | |
| "schedule": "sway", | |
| "source": "EN-E1 protocol", | |
| "steps": 32, | |
| "sway": -1.0, | |
| "w_speaker": 3.5, | |
| "w_text": 2.5 | |
| } | |
| }, | |
| "source": "en_grow_q/export/model_step_00000300.pt", | |
| "step": 300, | |
| "train_sources": [ | |
| "echo_en" | |
| ], | |
| "train_steps": 200000, | |
| "watermark": { | |
| "alpha": 1.0, | |
| "detector": "detector_base.pth", | |
| "generator": "generator_base.pth", | |
| "message": "0100110101010100", | |
| "method": "audioseal", | |
| "name": "audioseal_wm_16bits", | |
| "nbits": 16, | |
| "repo": "facebook/audioseal", | |
| "revision": "3c19eba53390776cf2cc9ed5f6c9ac67ce72ecba", | |
| "sample_rate": 16000 | |
| } | |
| } | |
| }, | |
| "policy": { | |
| "final_only": true, | |
| "main_min_step": 20000, | |
| "min_step": 20000 | |
| }, | |
| "prompts": { | |
| "01-short": "5d1822e59071fe22d6910d08f4c2c49735853be3cf50916b1e7cef17c246f477", | |
| "02-question": "ec5010cbb72864e62cd95a87840c6f74a750f4ac135ddcda43ce34417b63c346", | |
| "03-numbers": "4f1d94263fbe54bdcd239706933b97365dac4d34ea62f75e319da06821cdeff7", | |
| "04-conversational": "b7c93eb3bc3eb63828440322c7c79ec38df309d6476c5e6c7403b73576d4b031", | |
| "05-long": "9f6966023f8946ea8d9d915dbd3a8ac00c242217b131dcc979eed1063df2b9e0" | |
| }, | |
| "recipe": { | |
| "data.catalog": "catalogs/echo_en_v1", | |
| "data.max_frames": 10000, | |
| "data.p_cross": 0.6, | |
| "flow.t_mean": -0.8, | |
| "flow.t_std": 0.8, | |
| "source": "en_base/config.yaml", | |
| "train.decay_frac": 0.2, | |
| "train.grad_accum": 4, | |
| "train.hub_every": 10000, | |
| "train.lr": 0.00025, | |
| "train.max_steps": 200000, | |
| "train.schedule": "wsd", | |
| "train.warmup_steps": 5000 | |
| }, | |
| "run": "en_base", | |
| "validation": { | |
| "gt_utmos": { | |
| "echo_dev": 4.196, | |
| "seed_dev": 3.523 | |
| }, | |
| "pass": true, | |
| "summary": "Seed test-en GT WER 2.14 % (seed-tts-eval protocol; PASS), LS-PC GT WER 2.47 % (F5 protocol; PASS)" | |
| } | |
| } | |