Innovate Futures @ Benji
patreon
More Advanced MutliTalk Workflows - With TTS and Flux Lora
🕑 Added 2025-06-23 13:00:19 +0000 UTCVideo : https://youtu.be/koBHbr3IYpE
Related Post https://www.patreon.com/posts/132088744
Mentioned in the video, MultiTalk collaborating with other AI model such as Chatterbox for TTS , and Flux for character image.
Similar logic apply in Chatterbox and Fantasy Talking tutorials before.
Chatterbox TTS https://www.patreon.com/posts/comfyui-workflow-130376289
Fantasy Talking + Dia TTS + Flux https://www.patreon.com/posts/comfyui-workflow-127808329
Comments
Pronprom Pakpaim
I found error from node Download Wav2Vec Model DownloadAndLoadWav2VecModel expected str, bytes or os.PathLike object, not NoneType /transformers/models/wav2vec2/tokenization_wav2vec2.py", line 168, in __init__ with open(vocab_file, encoding="utf-8") as vocab_handle: TypeError: expected str, bytes or os.PathLike object, not NoneType # ComfyUI Error Report ## Error Details - **Node ID:** 137 - **Node Type:** DownloadAndLoadWav2VecModel - **Exception Type:** TypeError - **Exception Message:** expected str, bytes or os.PathLike object, not NoneType ## Stack Trace ``` ## System Information - **ComfyUI Version:** 0.3.43 - **Arguments:** main.py --listen 0.0.0.0 --port 8910 --auto-launch - **OS:** posix - **Python Version:** 3.10.18 (main, Jun 5 2025, 13:14:17) [GCC 11.2.0] - **Embedded Python:** false - **PyTorch Version:** 2.7.1+cu128 ## Devices - **Name:** cuda:0 NVIDIA GeForce RTX 5080 : cudaMallocAsync - **Type:** cuda - **VRAM Total:** 16609837056 - **VRAM Free:** 16329539584 - **Torch VRAM Total:** 0 - **Torch VRAM Free:** 0 Can help me fix this ?
현민 안
Hi. I updated ComfyUI, but now the workflow won't run. I was told to downgrade transformers to version 4.53.0, but even after downgrading, the issue still isn't resolved I think I might need to downgrade the version inside the python_embeded folder. What would be the correct way to do that https://github.com/kijai/ComfyUI-WanVideoWrapper/issues/690
Benjamin Law
Use MultiTalk
Scott Ownbey
which workflwo do you recommend> Fantasy Talking + Dia TTS + Flux or wanvideo_multitalk_FluxLoraCharacter?
Benjamin Law
Currently, this model do not have GGUF
Perilla(들깨)
Hi~ I'm always looking for help. Is it possible to make a WAN GGUF workflow? Can you help if possible?
Benjamin Law
Thats what i am testing tonight. You have join is the Discord? I can send over there.
macmotu
I have another question: Do you think, it is possible to get some sort of v2v workflow with multitalk impementation? Perhaps via VACE? I really would like to have some more movement, while lipsyncing. Prompting "She walks towards the camera" has no effect right now. I wonder whether the workflow can be extended or modified.
Benjamin Law
Hey Mac, Yes I understand what you mean. The facial expressions in this framework are most relay on the audio sound. So its good for conversational audio. If theres some big wave, it will project on the face. With prompt, I have try scene for example , happy face , in prompt. It is overreacting evil smile. Haha So I use prompt to control hand movements and body language expression only with prompt. Thats what I find.
macmotu
Hey Benji! I have tried your workflow and it is really great, but I am faced with a question: Is there a way to control the intensity of the facial expressions? Even if I enter something like ‘calm expression’ in the prompt, the result is always overacting. Presumably it is possible and I'm just too stupid to see it ;-)
