{
  "number": 18,
  "title": "Refined voice cloning settings",
  "body": "Update settings for voice cloning.",
  "state": "MERGED",
  "url": "https://github.com/potion/potion-voice/pull/18",
  "createdAt": "2023-04-21T05:10:41Z",
  "mergedAt": "2023-06-21T10:58:16Z",
  "closedAt": "2023-06-21T10:58:17Z",
  "additions": 1152,
  "deletions": 146,
  "changedFiles": 15,
  "isDraft": false,
  "baseRefName": "staging",
  "headRefName": "develop",
  "author": {
    "login": "author_unknown"
  },
  "mergedBy": {
    "login": "author_7"
  },
  "mergeCommit": {
    "oid": "54efcabc34b82b156ab06a0beab0f895d4c7edd0"
  },
  "milestone": null,
  "labels": {
    "nodes": []
  },
  "assignees": {
    "nodes": []
  },
  "requestedReviewers": {
    "nodes": []
  },
  "commits": {
    "totalCount": 32,
    "nodes": [
      {
        "commit": {
          "oid": "a47c5e000965b8c38d742a512d88c20d457cfd6a",
          "message": "Refined checkpointing and enabled weighted sampler for multi-speaker baseline training.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T17:55:31Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T17:55:31Z"
          }
        }
      },
      {
        "commit": {
          "oid": "c58120f853b1c7549818d7f7194d6fb901ebbc39",
          "message": "Refined checkpointing and enabled weighted sampler for multi-speaker baseline training.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T17:55:31Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T17:56:49Z"
          }
        }
      },
      {
        "commit": {
          "oid": "c53d9e44c068880953046b1ceab8e6da82b7f60c",
          "message": "Merge branch 'develop' of https://github.com/potion/potion-voice into develop",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T18:02:10Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-02-03T18:02:10Z"
          }
        }
      },
      {
        "commit": {
          "oid": "237c552fb6dafcf22c2213e69692cd22d6df43e1",
          "message": "Merge branch 'staging' into develop",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-21T05:13:59Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-21T05:13:59Z"
          }
        }
      },
      {
        "commit": {
          "oid": "452d734aefe8407bc0b018fa1974acbe1429dd05",
          "message": "Added support for Potion Diverse and Mozilla Common Voice data-sets as well as for continuation of training runs.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-27T02:37:20Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-27T02:37:20Z"
          }
        }
      },
      {
        "commit": {
          "oid": "c811784a9e37b78698f913bc63c76868e14b902f",
          "message": "Merge conflict resolved (naming convention diversion addressed.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-27T02:45:15Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-04-27T02:45:15Z"
          }
        }
      },
      {
        "commit": {
          "oid": "940e2d1af624e35dec9447b91495a2908c95c7b5",
          "message": "Test sentences added based on dataset arguments; training parameters revised; phoneme usage supported as argument.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-05T15:35:00Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-05T15:35:00Z"
          }
        }
      },
      {
        "commit": {
          "oid": "d6dea65f95cc3a8d8f3e37f62cc022f908a1a936",
          "message": "Add loss to config explicitely; use different resblock_type_decoder by default",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-09T07:09:26Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-09T07:09:26Z"
          }
        }
      },
      {
        "commit": {
          "oid": "4a9ddf2b561bf6240d3d03da8671e77b7e7ed78a",
          "message": "phoneme support; updated training parameters; removed use_cpu option (unsupported).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-15T07:48:30Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-15T07:48:30Z"
          }
        }
      },
      {
        "commit": {
          "oid": "2563f8dee714cd8b83e72a511c37c7daa4db5c19",
          "message": "Remove overwrite of phoneme usage; use setting from config.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-15T16:07:22Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-15T16:07:22Z"
          }
        }
      },
      {
        "commit": {
          "oid": "31dd16cc50d27f434b0afa719f44ad4565a9f17b",
          "message": "Added new script to generate a merged speaker embeddings file (for multiple datasets).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-17T16:14:46Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-17T16:14:46Z"
          }
        }
      },
      {
        "commit": {
          "oid": "96680990014cc7c8943fdec303e04299ec5a5914",
          "message": "Minor bug fix (argument misspelled).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-17T16:24:30Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-17T16:24:30Z"
          }
        }
      },
      {
        "commit": {
          "oid": "a70ed6dc7f33de96b19390bfa66a2e755e899232",
          "message": "Removed unnecessary import; improved comments.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T02:27:06Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T02:27:06Z"
          }
        }
      },
      {
        "commit": {
          "oid": "f0a7f3cfa96afbb14defc01c2484559d13a9eb2a",
          "message": "Added new voice conversion option (idea 1: convert from multi-speaker model; idea 2: clone voice, synthesize speech, convert into recorded template; ...).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T15:39:33Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T15:39:33Z"
          }
        }
      },
      {
        "commit": {
          "oid": "abb23d49febcb47434f8d5e4db467f86bdd242cb",
          "message": "Replaces hard-coded spk embedding reference; commented code more thoroughly.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T16:45:58Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-18T16:45:58Z"
          }
        }
      },
      {
        "commit": {
          "oid": "2ed529263af58b0c8374306bd530ab58892df2e7",
          "message": "Added new voice cloning approach (using fine-tuning via continue_path).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T08:53:49Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T08:53:49Z"
          }
        }
      },
      {
        "commit": {
          "oid": "b3d7d633595b26c75c34c6e876685ae0e3008977",
          "message": "Spelling and formatting improvements.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T08:54:37Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T08:54:37Z"
          }
        }
      },
      {
        "commit": {
          "oid": "7836c1cbf74f408d76bd2777ff38dc116b596951",
          "message": "Revised training parameters after testing (no freeze and unique phoneme cache per run).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T10:10:03Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T10:10:03Z"
          }
        }
      },
      {
        "commit": {
          "oid": "3f80bdcb2a4efdb4c7086c8e4985eca28d0262c8",
          "message": "Fixed typo",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T10:13:20Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-19T10:13:20Z"
          }
        }
      },
      {
        "commit": {
          "oid": "d59d33fb0190c1b42148804e112e8c3578544fbb",
          "message": "Explicitly name phonemizer (i.e., espeak).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T02:25:36Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T02:25:36Z"
          }
        }
      },
      {
        "commit": {
          "oid": "913e45d5b8d16859d74d1ae327835473ee7d4d8c",
          "message": "Remove scaling factors for now; improved conversion settings.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T14:57:32Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T14:57:32Z"
          }
        }
      },
      {
        "commit": {
          "oid": "320b26f304a147b511d015487de6dcc62d4d17d9",
          "message": "Added numpy import for assert statements.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T15:03:21Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-22T15:03:21Z"
          }
        }
      },
      {
        "commit": {
          "oid": "4ad5caa4edc3fd79c4fddc52f91fd7b8d67902ef",
          "message": "Fixed mistake in target_auddio_embedding computation.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-23T07:48:57Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-23T07:48:57Z"
          }
        }
      },
      {
        "commit": {
          "oid": "2bf1db90a774abebce8f68cb92428807a0fd2c33",
          "message": "Added clean-up steps to remove temporary files.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-23T15:01:35Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-23T15:01:35Z"
          }
        }
      },
      {
        "commit": {
          "oid": "5ae957485632e7e49295104ff9c7095ab3ec11d2",
          "message": "Fixed typo.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:44:19Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:44:19Z"
          }
        }
      },
      {
        "commit": {
          "oid": "87aeefc7689f0cb76a0cc8455178015be11fb04e",
          "message": "Added (but left them commented out) additional configuration options used in the YourTTS paper.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:45:21Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:45:21Z"
          }
        }
      },
      {
        "commit": {
          "oid": "0b29c3b698a063048762c42c90b6f653371e9135",
          "message": "Added scale related synth parameters (to be tested / refined).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:56:14Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T02:56:14Z"
          }
        }
      },
      {
        "commit": {
          "oid": "e23cc5d5abb6d6f87d55606088e05edd6e15021f",
          "message": "Added sampiling rate as argument (in preparation for higher quality training runs).",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T15:02:21Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-24T15:02:21Z"
          }
        }
      },
      {
        "commit": {
          "oid": "1e036850c9c3514acfce3ab549ad1a7953b37ca2",
          "message": "Added ffmpeg-normalize and require resemblyzer>=1.3.0.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-25T16:57:17Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-25T16:57:17Z"
          }
        }
      },
      {
        "commit": {
          "oid": "add8aec51cb065024e3658ba3abd24cddb3af0e4",
          "message": "New cloning step and deployment guidance added; TTS requirements updated.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-25T16:58:39Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-05-25T16:58:39Z"
          }
        }
      },
      {
        "commit": {
          "oid": "397d59ea1071383b5cf94b584028f5040d83f951",
          "message": "Fixed typo in requirements files.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-06-21T10:03:06Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-06-21T10:03:06Z"
          }
        }
      },
      {
        "commit": {
          "oid": "08d6314e0270b92a1dcab69b30ba57313badd7ea",
          "message": "TTS repo assumes config.json name is fixed ... it's hardcoded; so, keep it as config.json.",
          "author": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-06-21T10:22:39Z"
          },
          "committer": {
            "name": "author_unknown",
            "email": "author_unknown",
            "date": "2023-06-21T10:22:39Z"
          }
        }
      }
    ]
  },
  "reviews": {
    "nodes": []
  },
  "comments": {
    "nodes": [
      {
        "author": {
          "login": "author_unknown"
        },
        "body": "@author_7 @author_6 This is ready to be merged into staging!",
        "createdAt": "2023-05-25T17:05:27Z",
        "url": "https://github.com/potion/potion-voice/pull/18#issuecomment-1563235903"
      }
    ]
  },
  "files": {
    "nodes": [
      {
        "path": "requirements.dev.local.txt",
        "additions": 2,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "requirements.dev.txt",
        "additions": 2,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "requirements.prod.cpu.txt",
        "additions": 2,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "requirements.prod.gpu.txt",
        "additions": 2,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "requirements.txt",
        "additions": 2,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/clone_voice.py",
        "additions": 17,
        "deletions": 22,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/clone_voice_via_continue.py",
        "additions": 278,
        "deletions": 0,
        "changeType": "ADDED"
      },
      {
        "path": "voice-cloning/convert_voice.py",
        "additions": 308,
        "deletions": 0,
        "changeType": "ADDED"
      },
      {
        "path": "voice-cloning/docs/potion-voice-cloning_Installation_Guide.md",
        "additions": 153,
        "deletions": 4,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/prepare_datasets.py",
        "additions": 84,
        "deletions": 1,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/save_multispeaker_baseline_embeddings_file.py",
        "additions": 107,
        "deletions": 0,
        "changeType": "ADDED"
      },
      {
        "path": "voice-cloning/synthesize_speech.py",
        "additions": 5,
        "deletions": 0,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/train_config.py",
        "additions": 21,
        "deletions": 2,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/train_multispeaker_baseline_model.py",
        "additions": 168,
        "deletions": 108,
        "changeType": "MODIFIED"
      },
      {
        "path": "voice-cloning/utils/synthesize_utils.py",
        "additions": 1,
        "deletions": 4,
        "changeType": "MODIFIED"
      }
    ]
  }
}