| | --- |
| | license: agpl-3.0 |
| | datasets: |
| | - hackersgame/hamner |
| | tags: |
| | - Hamner |
| | - Purism |
| | - voice |
| | --- |
| | # Hamner Voice |
| |
|
| | Just a voice for MeloTTS. This a result of following [this](https://github.com/myshell-ai/MeloTTS/blob/main/docs/training.md). |
| |
|
| | ## Model Details |
| | This was run for only 12 hours on a single 3090. I'm sure you could get a lot better results running for longer/on more hardware. |
| |
|
| | ## Uses |
| |
|
| | This lets you make audio in my voice (David Hamner). You can also train your own model with this [data set](https://huggingface.co/datasets/hackersgame/hamner) |
| |
|
| | ### Direct Use |
| |
|
| | Setup [this](https://github.com/myshell-ai/MeloTTS/). |
| | Copy the pth file from this repo and run: `python infer.py --text "Hey there, David Hamner here, and now you can use my voice! As always it's GPL!" -m ./logs/out/G_59000.pth -o ./test` |
| |
|
| | ### Out-of-Scope Use |
| |
|
| | Don't misuse this. Don't impersonate me or Purism, always note an AI voice was used. |
| |
|
| |
|
| | ### Training Data |
| |
|
| | [data set](https://huggingface.co/datasets/hackersgame/hamner) |
| |
|
| | ### Training Procedure |
| |
|
| | [this](https://github.com/myshell-ai/MeloTTS/blob/main/docs/training.md) |
| |
|
| |
|
| | ### Results |
| |
|
| | [Example](output.wav) |
| |
|
| |
|
| | ## Model Card Authors |
| |
|
| | David Hamner |
| | Based on: https://github.com/myshell-ai/MeloTTS |
| |
|
| | ## Model Card Contact |
| |
|
| | [David Hamner](https://www.linkedin.com/in/david-hamner-a1633635/) |
