TTS server and xtts not works with fine tuned models (#3476) #3544
Replies: 4 comments 9 replies
-
|
Currently attempting to run a fine tuned xtts v2 model like (with these changes applied: bivashy:fix/xtts-server-error) : and
Neither works and returns this error: |
Beta Was this translation helpful? Give feedback.
-
|
Here are readable format of the error: Traceback (most recent call last):
File "/usr/local/bin/tts-server", line 33, in <module>
sys.exit(load_entry_point('TTS', 'console_scripts', 'tts-server')())
File "/usr/local/bin/tts-server", line 25, in importlib_load_entry_point
return next(matches).load()
File "/usr/lib/python3.10/importlib/metadata/__init__.py", line 171, in load
module = import_module(match.group('module'))
File "/usr/lib/python3.10/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1050, in _gcd_import
File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 688, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 883, in exec_module
File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
File "/root/TTS/server/server.py", line 104, in <module>
synthesizer = Synthesizer(
File "/root/TTS/utils/synthesizer.py", line 93, in __init__
self._load_tts(tts_checkpoint, tts_config_path, use_cuda)
File "/root/TTS/utils/synthesizer.py", line 183, in _load_tts
self.tts_config = load_config(tts_config_path)
File "/root/TTS/config/__init__.py", line 97, in load_config
config_class = register_config(model_name.lower())
File "/root/TTS/config/__init__.py", line 40, in register_config
from TTS.tts.configs.xtts_config import XttsConfig
File "/root/TTS/tts/configs/xtts_config.py", line 5, in <module>
from TTS.tts.models.xtts import XttsArgs, XttsAudioConfig
File "/root/TTS/tts/models/xtts.py", line 67, in <module>
manager = ModelManager(path)
File "/root/TTS/utils/manage.py", line 56, in __init__
self.read_models_file(models_file)
File "/root/TTS/utils/manage.py", line 68, in read_models_file
self.models_dict = read_json_with_comments(file_path)
File "/root/TTS/config/__init__.py", line 17, in read_json_with_comments
with fsspec.open(json_path, "r", encoding="utf-8") as f:
File "/usr/local/lib/python3.10/dist-packages/fsspec/core.py", line 103, in __enter__
f = self.fs.open(self.path, mode=mode)
File "/usr/local/lib/python3.10/dist-packages/fsspec/spec.py", line 1295, in open
f = self._open(
File "/usr/local/lib/python3.10/dist-packages/fsspec/implementations/local.py", line 180, in _open
return LocalFileOpener(path, mode, fs=self, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/fsspec/implementations/local.py", line 302, in __init__
self._open()
File "/usr/local/lib/python3.10/dist-packages/fsspec/implementations/local.py", line 307, in _open
self.f = open(self.path, mode=self.mode)
FileNotFoundError: [Errno 2] No such file or directory: '/root/TTS/tts/models/../.models.json'For some reason in Maybe you've downloaded/using modified tts? |
Beta Was this translation helpful? Give feedback.
-
|
I'm using the docker container: and modified the files you've pushed |
Beta Was this translation helpful? Give feedback.
-
|
Wondering if anyone else is running into this issue? |
Beta Was this translation helpful? Give feedback.
-
|
I am sorry, currently i have no idea |
Beta Was this translation helpful? Give feedback.
-
|
OK so, I replaced the files included in the docker with the latest release (I also used your modifications @bivashy ) and I get a different error now. I've changed how I'm running it to the following: the result is the following:
|
Beta Was this translation helpful? Give feedback.
-
|
OK, so I finally got this working. I had to modify /root/TTS/utils/synthesizer.py with the following
Now I'm able to generate TTS via the web page. The main usage for me was to link this up with home assistant (Mary TTS integration). I'll test that today to make sure everything works correctly. |
Beta Was this translation helpful? Give feedback.
-
|
OK, I needed to make some changes in order to have the default web app AND for Home Assistant to work properly. I've confirmed that these changes work with both the included webapp and via MaryTTS in Home Assistant (using the docker container provided by coqui-tts) |
Beta Was this translation helpful? Give feedback.
-
|
hello @ther3zz , I hope you doing great. I was hoping you could give me some insight. I don't want to put too much info here just because I was trying for a hole day so I have a lot to say. But, basically I have hass, I made a custom voice that work with piper just fine but I found it way to robotic. Then, I found out about XTTS and I was mindblown by the result. I want that. I am unable to make it work. I was able to have the TTS server running, even integrated it in my compose file for my hole AI-stack. My main issue seams to be that I have made a multi-speaker model based on the logs/console, and I need to specify somehow the speaker. I used this to create/fine-tune the voice. xtts-finetune-webui. I realy don't get how to make it work. And I tried. I also gave a shot at your modified files and line you mentioned, cloned your repo. Tried the maintained fork, the old one. But, it always result into "not working" because I change some .py files get some error or because it cry out about my model file speaker thingy. Anyway... I have more prescise info to share that is for sure like logs etc. But I'll wait for your reply or someone else for now. Thank you ! Edit: ended up using something else for hosting my xtts model. |
Beta Was this translation helpful? Give feedback.
-
|
Ther3zz.... @ther3zz I am running similar issues as you running Coqui on docker with the fine tuned models. I am using them for Home Automation to Google minis around the house. I got these voices working using the server..... https://huggingface.co/enlyth/baj-tts/tree/main and they are great. I am trying some others such as Bob Odenkirk (better call Saul)... I am having problems with all the Fined Tune models. https://huggingface.co/drewThomasson/xtts-finetune-Bob-Odenkirk/tree/main I tried making your changes within my container and had no luck... Doesn't appear to break anything.... I am wondering if you could push your whole container.... I am using the same container as you... image: ghcr.io/coqui-ai/tts-cpu:latest Though I might be using the CPU version where you are using the GPU. |
Beta Was this translation helpful? Give feedback.
-
I ended up using the openedai-speech (https://github.com/matatonic/openedai-speech) project over the xtts server from coqui It's WAY better |
Beta Was this translation helpful? Give feedback.
-
I forgot to mention, if you're using home assistant, you'll need to use something like this in hacs: https://github.com/sfortis/openai_tts |
Beta Was this translation helpful? Give feedback.
-
|
Our fork now supports all Coqui TTS models in the server. Specifying a |
Beta Was this translation helpful? Give feedback.
-
Question: #3476 (comment)
Beta Was this translation helpful? Give feedback.
All reactions