Budget: 4000 UAH Deadline: 2 days
Hello! I will set up the F5-TTS/Misha24-10 (or Chatterbox) workflow in your ComfyUI under StabilityMatrix for generating long MP3 audiobooks from text. I have experience with ComfyUI pipelines and local AI tuning in Python. I work via remote access (AnyDesk/TeamViewer) to your PC, and we can agree on a convenient time in writing. The cost and timeline are below.
Budget: 4000 UAH Deadline: 2 days
Good day! I set up the ComfyUI workflow locally, including F5-TTS — it works well for long texts in audiobooks. Since the StabilityMatrix with ComfyUI is already up, we just need to install the custom nodes for F5-TTS and build the workflow for batch voiceover from txt to MP3. I will connect via AnyDesk, install it, and run it using your example in about two days. If F5 doesn't meet the voice requirements, I can also set up Chatterbox.
Budget: 4000 UAH Deadline: 1 day
Good day, Igor. What is the average volume of text that you plan to upload at one time? (This is needed so I can properly configure the node for automatic text splitting into chunks; otherwise, the neural network will overload the video memory). What model of graphics card is installed on the PC? (F5-TTS works quickly, but the batch sizes need to be optimized for your VRAM capacity). Should the final audiobooks be compiled into one monolithic MP3 file, or should the workflow break them down by chapters?
I can provide the assembly to you as a ready JSON file of the workflow with text instructions, or I can connect via AnyDesk and set everything up, including installing the necessary Python dependencies (vocos, transformers) directly in the StabilityMatrix environment.
I would be happy to help.
Budget: 4000 UAH Deadline: 2 days
Need to set up the F5-TTS workflow in the already functioning ComfyUI and obtain MP3 output from text files of acceptable quality.
Steps: I will install the custom F5-TTS node (Misha24-10 or Chatterbox) via ComfyUI Manager, select a working workflow for standard voices without cloning, configure batch processing of long texts with chunking and merging into the final MP3. The main risk point here is the Python dependencies and the version of torch: F5-TTS often conflicts with the ComfyUI environment, which can be resolved through a separate venv or by editing requirements.
What is your graphics card and how much VRAM do you have -- this will determine which model can be used without OOM and how fast the generation will be?
Budget: 4000 UAH Deadline: 2 days
Good day!
I am interested in your project. I am ready to help with the setup of F5-TTS (or an alternative solution if it works more stably) in the already deployed Stability Matrix + ComfyUI. I will check the workflow, set up the generation of long audio files from text, and help resolve any potential configuration issues.
To assess, I would like to clarify a few details:
what graphics card do you have and how much VRAM;
what version of ComfyUI is being used;
is remote access setup planned;
is it essential to use F5-TTS specifically, or can a similar solution be applied if it provides better stability?
After clarifying the details, I am ready to start working immediately.
I would be happy to collaborate. If anything, feel free to message me privately, we will work it out.
Budget: 4000 UAH Deadline: 3 days
Good day! We have experience working with ComfyUI and setting up local TTS solutions. We will implement F5-TTS integration for batch processing of text files into audiobooks by creating a custom workflow that will ensure stable generation of long MP3s without failures. We will configure all necessary dependencies and output parameters for proper operation on your equipment.
