{ "number": 23, "title": "Develop", "body": "Support of 48k Hz sampling rate as default; GPU and CPU-based usage for all scripts but baseline training run.\r\n\r\nSee voice-cloning/docs/Voice\\ Cloning\\ @\\ 48k\\ Hz\\ Sampling\\ Rate\\ -\\ Step-by-Step.txt for example usage.\r\n\r\nModel assets can be found at s3://potion-ai-models/potion-voice-2023-08/", "state": "MERGED", "url": "https://github.com/potion/potion-voice/pull/23", "createdAt": "2023-08-07T06:58:31Z", "mergedAt": "2023-09-25T07:15:56Z", "closedAt": "2023-09-25T07:15:56Z", "additions": 4121, "deletions": 180, "changedFiles": 27, "isDraft": false, "baseRefName": "staging", "headRefName": "develop", "author": { "login": "author_unknown" }, "mergedBy": { "login": "author_7" }, "mergeCommit": { "oid": "8d62d38e524bfc62be0b6ac0ebca155ec2c4dda7" }, "milestone": null, "labels": { "nodes": [] }, "assignees": { "nodes": [] }, "requestedReviewers": { "nodes": [] }, "commits": { "totalCount": 46, "nodes": [ { "commit": { "oid": "6caecc390d8da5c2a0e1028ef5a477a25d0af9b5", "message": "Switch to 48k as default sampling rate; disable mixed_precision due to possible min()/max() runtime error.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-06-30T11:29:26Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-06-30T11:29:26Z" } } }, { "commit": { "oid": "f456697d843e83a34b1b5c192ced548bd10ae9d5", "message": "Add progress tracker and suppress default TTS command-line outputs.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-17T08:32:37Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-17T08:32:37Z" } } }, { "commit": { "oid": "dacf5b23d8581e816104868923dfc5d872960e5e", "message": "Add progress tracker and suppress default TTS command-line outputs (part 2).", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-18T15:24:55Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-18T15:24:55Z" } } }, { "commit": { "oid": "a6639dbbae3c7920b312b5077aea4a9b42c18e1e", "message": "Merge pull request #21 from potion/dev-find-best-fixup\n\nDev find best fixup", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-21T05:14:54Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-21T05:14:54Z" } } }, { "commit": { "oid": "31234863f06285dcb88732cff5eead330ab852ad", "message": "Minor improvements to support CPU-only processing.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T07:52:47Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T07:52:47Z" } } }, { "commit": { "oid": "c96dccfb58c49186ee21c4fb24eff96302b35ecf", "message": "Added more usage examples and necessary Coqui.ai TTS changes.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T09:43:09Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T09:43:09Z" } } }, { "commit": { "oid": "19894f8d753f8b5ecf7fb8ee756b63c4498f0fe9", "message": "Merge pull request #22 from potion/dev-cpu-only\n\nDev cpu only", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T09:44:26Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-25T09:44:26Z" } } }, { "commit": { "oid": "2fde687781a036630ad65a1b6438f2bd2fe860c8", "message": "Updated requirements: git clone --depth 1 --branch v0.16.0 https://github.com/coqui-ai/TTS onwards addresses the CPU-only processing issues.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T07:45:02Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T07:45:02Z" } } }, { "commit": { "oid": "5f74690d758f047d1aa24f03b72dd242cfdf9c14", "message": "Efficiency improvements: Rearranged loop order and removed re-init of speaker mgr and vocoder.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:42:56Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:42:56Z" } } }, { "commit": { "oid": "4b83dcd2b33940fb67aa0a2ea726e5f4825709af", "message": "Moved tqdm to outermost loop.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:49:29Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:49:29Z" } } }, { "commit": { "oid": "ff2efad2181eef60c95855a26a346db3a241ac33", "message": "Efficiency improvement: Removed re-init of vocoder.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:58:54Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-26T16:58:54Z" } } }, { "commit": { "oid": "bfab4e52d152e1c1adb5127d8c52e57e5926d6f2", "message": "Enable use_speaker_encoder_as_loss byu default.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-27T08:45:46Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-27T08:45:46Z" } } }, { "commit": { "oid": "c4f0738615226b541c2728d8b518c1bb8f422e59", "message": "Disable use_speaker_encoder_as_loss until we have a 48k speaker encoder.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-27T10:14:48Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-07-27T10:14:48Z" } } }, { "commit": { "oid": "ebc3aabb5b47b92667f5cf38c6ad3eabe927c29b", "message": "Efficiency improvements: re-use config and speaker embeddings when multiple models are evaluated.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-01T05:19:00Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-01T05:19:00Z" } } }, { "commit": { "oid": "efe67a3be7ab363098a6ac11a868c84db8401942", "message": "Bug fix: removed unnecessary argument.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-01T05:51:24Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-01T05:51:24Z" } } }, { "commit": { "oid": "9c25385822430a0be1b517eba819fe45598b9a22", "message": "No more need to suppress cmd output; fix remove silence bug; code readability improvements; better error handling.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T04:12:11Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T04:12:11Z" } } }, { "commit": { "oid": "b861f6e7ca39117ce9d9d115dea10efd0b9382ad", "message": "Fixed Vits config & model imports / typing references.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T05:20:28Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T05:20:28Z" } } }, { "commit": { "oid": "24db1203681e633095501ad987fb7de61acbe1bf", "message": "Fixed error handling for speaker embeddings.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T05:28:06Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T05:28:06Z" } } }, { "commit": { "oid": "a9aa70cd38e28c26b9ea9cf3a142607733b5355e", "message": "Typing improvements; trim_silence improvements.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T06:58:39Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T06:58:39Z" } } }, { "commit": { "oid": "5a9113804a0762d4c926a5df9b56cd9cde09681d", "message": "Testing alternative sim_score routine.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T09:10:37Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T09:10:37Z" } } }, { "commit": { "oid": "9d937c7020d4fb2f71873beac200e12db77e2863", "message": "Improved speaker similarity scoring approach.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T09:59:09Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T09:59:09Z" } } }, { "commit": { "oid": "e81db101ed94eca0a634b16bbe0dc27f40ceb81e", "message": "Remove scoring test import.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T10:05:18Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T10:05:18Z" } } }, { "commit": { "oid": "5b551ce70ad459b64a9cdd2ce491a8d7740c44b0", "message": "Fix ValueError issue.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T11:09:13Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T11:09:13Z" } } }, { "commit": { "oid": "77fb20ca06b2f8b76285d6eefe662def5c92301b", "message": "Add unload model (clean-up) routine; code readibility improvements.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T16:54:23Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-02T16:54:23Z" } } }, { "commit": { "oid": "afdf4cfd7cc0cf90710a2e602327db45d6a87088", "message": "Add unload model (clean-up) routine.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-03T01:42:31Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-03T01:42:31Z" } } }, { "commit": { "oid": "60c8b4fe11e23740c4887cd8b5740b49eb63c806", "message": "Switching to 48k HZ sampling rate by default.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T01:07:45Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T01:07:45Z" } } }, { "commit": { "oid": "e47841f80c1e73ac33abbc161504782c3b06959b", "message": "sampling_rate check fixes", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T01:28:27Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T01:28:27Z" } } }, { "commit": { "oid": "f98ad3fa51ed45847377d0d648c63868c2c990c6", "message": "Support minimise call for multiple models (i.e., skip saving / overwritting minimised config).", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T04:55:43Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T04:55:43Z" } } }, { "commit": { "oid": "a4d002539ad67e0667e6642293123793d7b72a34", "message": "Refreshed documentation to reflect 48k sampling rate setup and usage.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T06:52:42Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T06:52:42Z" } } }, { "commit": { "oid": "a92f32ac694e18e19b2676847387a215d05e01ea", "message": "Merge branch 'staging' into develop-07-08", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:03:54Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:03:54Z" } } }, { "commit": { "oid": "1bd4b8d3039d8a1e2ac02476bec235da829c5b5c", "message": "Merge branch 'staging' into develop-07-08", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:05:02Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:05:02Z" } } }, { "commit": { "oid": "75e78312e8d92560c335f640248bef88e37ddcf7", "message": "Merge pull request #24 from potion/develop-07-08\n\nDevelop 07 08 code updates", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:08:05Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:08:05Z" } } }, { "commit": { "oid": "a3afe89e0e4289c33044e3b9401a5d0bda2be401", "message": "Added code for cloning script", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:20:41Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:20:41Z" } } }, { "commit": { "oid": "1c59012333173eba76cec6325608024beb1a669a", "message": "Merge pull request #25 from potion/develop-07-08-updates\n\nAdded code for cloning script", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:21:18Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T08:21:18Z" } } }, { "commit": { "oid": "ea4608475587cea4615a4980bc767708a186d57e", "message": "Added changes for sr 48000", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T10:17:46Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T10:17:58Z" } } }, { "commit": { "oid": "6defbb0bb874390f1b1b51b1e20355890c530493", "message": "Merge pull request #26 from potion/develop-07-08-updates\n\nAdded changes for sr 48000", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T10:19:55Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-08-07T10:19:55Z" } } }, { "commit": { "oid": "bb152205ade80927946b88008109f106107af8f9", "message": "Extended similarity scoring approach to also includde naturalness assessment.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-13T16:09:30Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-13T16:09:30Z" } } }, { "commit": { "oid": "34740bd2d5daf5411c3fd51dd81125c3e2ee2a12", "message": "Merge branch 'develop' of https://github.com/potion/potion-voice into develop", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-13T16:11:05Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-13T16:11:05Z" } } }, { "commit": { "oid": "9527d33af85af9dae620cdf2fb6e462c50c9979e", "message": "Bug fix: Wrong path used for synth samples.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T01:04:16Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T01:04:16Z" } } }, { "commit": { "oid": "49a20a14ab518f005305c22e14bdfe4efab2d7cf", "message": "Added naturalness score to find_best_cloned_model; improved comments.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T07:52:52Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T07:52:52Z" } } }, { "commit": { "oid": "9703a12253c7e870079654776d8458a06725c71c", "message": "Bug fix: init quality score correction", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T07:55:21Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T07:55:21Z" } } }, { "commit": { "oid": "1021bd8bd93694350d66caf9ee7e6c03d2e4b1d9", "message": "voice quality score redefined: 3/4 sim_scoe and 1/4 nat_score.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T14:34:46Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-14T14:34:46Z" } } }, { "commit": { "oid": "789d1bc6777f274153b00a071e62cee7b689ae40", "message": "Readying updated scoring approach for staging.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:18:03Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:18:03Z" } } }, { "commit": { "oid": "313d2bda3895b0de175b9692d0148c0f257d3f1c", "message": "Bug fix: Adjust to prev name change of init_synth argument.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:23:51Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:23:51Z" } } }, { "commit": { "oid": "5233e2e6e3655d3f5f77dd1abf23a482d7c3975e", "message": "Init fixup; improved comments.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:52:22Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T05:52:22Z" } } }, { "commit": { "oid": "acbc3c99460ab52b1d5e6cd8bbf92dfe1062fdfe", "message": "Improved wording for console-based output.", "author": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T07:40:39Z" }, "committer": { "name": "author_unknown", "email": "author_unknown", "date": "2023-09-18T07:40:39Z" } } } ] }, "reviews": { "nodes": [] }, "comments": { "nodes": [] }, "files": { "nodes": [ { "path": ".gitignore", "additions": 2, "deletions": 1, "changeType": "MODIFIED" }, { "path": "requirements.dev.local.txt", "additions": 6, "deletions": 0, "changeType": "MODIFIED" }, { "path": "requirements.dev.txt", "additions": 9, "deletions": 3, "changeType": "MODIFIED" }, { "path": "requirements.prod.cpu.txt", "additions": 9, "deletions": 4, "changeType": "MODIFIED" }, { "path": "requirements.prod.gpu.txt", "additions": 6, "deletions": 0, "changeType": "MODIFIED" }, { "path": "requirements.txt", "additions": 6, "deletions": 0, "changeType": "MODIFIED" }, { "path": "voice-cloning-job-handler/index.js", "additions": 3, "deletions": 3, "changeType": "MODIFIED" }, { "path": "voice-cloning/clone_voice.py", "additions": 14, "deletions": 12, "changeType": "MODIFIED" }, { "path": "voice-cloning/clone_voice_via_continue.py", "additions": 6, "deletions": 5, "changeType": "MODIFIED" }, { "path": "voice-cloning/docs/Voice Cloning @ 48k Hz Sampling Rate - Step-by-Step.txt", "additions": 183, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/docs/potion-voice-cloning_Installation_Guide.md", "additions": 20, "deletions": 6, "changeType": "MODIFIED" }, { "path": "voice-cloning/docs/potion-voice-cloning_Installation_Guide_-_CPU_only.md", "additions": 539, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/find_best_cloned_model.py", "additions": 58, "deletions": 24, "changeType": "MODIFIED" }, { "path": "voice-cloning/find_best_multispeaker_model.py", "additions": 77, "deletions": 37, "changeType": "MODIFIED" }, { "path": "voice-cloning/minimize_cloned_voice_model.py", "additions": 7, "deletions": 3, "changeType": "MODIFIED" }, { "path": "voice-cloning/prepare_datasets.py", "additions": 5, "deletions": 6, "changeType": "MODIFIED" }, { "path": "voice-cloning/score_cloned_voice.py", "additions": 18, "deletions": 10, "changeType": "MODIFIED" }, { "path": "voice-cloning/synthesize_speech.py", "additions": 29, "deletions": 18, "changeType": "MODIFIED" }, { "path": "voice-cloning/train_multispeaker_baseline_model.py", "additions": 32, "deletions": 24, "changeType": "MODIFIED" }, { "path": "voice-cloning/utils/NISQA/LICENSE", "additions": 21, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/NISQA/LICENSE_model_weights", "additions": 437, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/NISQA/NISQA_lib.py", "additions": 2170, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/NISQA/NISQA_model.py", "additions": 238, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/NISQA/nisqa_tts.tar", "additions": 0, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/misc_utils.py", "additions": 38, "deletions": 0, "changeType": "ADDED" }, { "path": "voice-cloning/utils/scoring_utils.py", "additions": 98, "deletions": 9, "changeType": "MODIFIED" }, { "path": "voice-cloning/utils/synthesize_utils.py", "additions": 90, "deletions": 15, "changeType": "MODIFIED" } ] } }