diff --git a/recipes/quickstart/NotebookLlama/README.md b/recipes/quickstart/NotebookLlama/README.md index dbaeab917a14ab37d3549ef80c78b6df2f223532..bb82fcabaab5e7b9060bd043bb4bb82a02a74265 100644 --- a/recipes/quickstart/NotebookLlama/README.md +++ b/recipes/quickstart/NotebookLlama/README.md @@ -20,6 +20,17 @@ Running Debator: ```python 1B-debating-script.py --initial_topic "The future of ### Scratch-pad/Running Notes: +Bark-Tests: Best results for speaker/v6 are at ```speech_output = model.generate(**inputs, temperature = 0.9, semantic_temperature = 0.8) +Audio(speech_output[0].cpu().numpy(), rate=sampling_rate)``` + +Tested sound effects: + +- Laugh is probably most effective +- Sigh is hit or miss +- Gasps doesn't work +- A singly hypen is effective +- Captilisation makes it louder + Ignore/Delete this in final stages, right now this is a "vibe-check" for TTS model(s): - https://github.com/SWivid/F5-TTS: Latest and most popular-"feels robotic"