nodes/BobRandomNumber-ComfyUI-DiaTTS
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
BobRandomNumber
2025-04-26 02:43:50 -04:00
committed by GitHub
parent 9ee960e77e
commit 0ed215e01a
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+4 -2
View File
@@ -1,6 +1,8 @@
# ComfyUI DiaTest TTS Node
This node pack integrates the [Nari Labs Dia](https://github.com/nari-labs/dia) text-to-speech model into ComfyUI using a single node for loading (onto GPU) and generation.
Warning LLM Code there are probably better options
This node pack partially integrates the [Nari Labs Dia](https://github.com/nari-labs/dia) text-to-speech model into ComfyUI using a single node for loading (onto GPU) and generation.
Dia allows generating dialogue with speaker tags (`[S1]`, `[S2]`) and non-verbal sounds (`(laughs)`, etc.). This node loads the model from Hugging Face Hub and generates audio directly using float32 precision. It requires a CUDA-enabled GPU.
@@ -57,4 +59,4 @@ Loads the specified Dia model from Hugging Face Hub onto the GPU (if not already
* The first time you run the node for a specific `repo_id`, it will download the model files from Hugging Face Hub, which may take some time. Subsequent runs will use the cached model.
* Changing the `repo_id` will trigger a model reload.
* The model uses float32 precision internally.
* The Descript Audio Codec (DAC) dependency (`descript-audio-codec`) must be installed via `requirements.txt`.
* The Descript Audio Codec (DAC) dependency (`descript-audio-codec`) must be installed via `requirements.txt`.