Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
v3ucn opened a pull request on SWivid/F5-TTS
Solve the problem of writing Chinese garbled code to csv training set…
When reading csv files, Excel identifies the encoding by reading the BOM on the file header. If the file header has no BOM information, it will read according to the Unicode encoding by default. Th...
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
ModuleNotFoundError: No module named 'encodec.model'
I am running on Windows. Using pyenv to set up a virtual environment and Python 3.10.11. I ran the two commands for installing pytorch and then ran the command to install requirements.txt. All...
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
对中文的使用拼音写的文字，转义效果不好
比如，“采用jieba分词工具进行分词”，中对“jieba“ TTS完全识别不到是中文的读法jieba， 二是使用英文读法了。 这个有什么方式改善吗？
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
TypeError: 'type' object is not subscriptable
Traceback (most recent call last): File "inference-cli.py", line 19, in <module> from model import CFM, DiT, MMDiT, UNetT File "/opt/SWivid/F5-TTS/F5-TTS/model/__init__.py", line 7, in <...
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
If I want to customize a model with high sound quality and similar timbre, do I need to fine tune the base model
If I want to customize a model with high sound quality and similar timbre, do I need to fine tune the base model
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
generation issues
1) using inference-cli with a large text splits the generation into multiple batches, but at the start of the next batch a random sounds gets added. 2) generation reads hyphenated compound word ...
SWivid created a comment on an issue on SWivid/F5-TTS
will close this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
Could this be used for singing ?
Hi, awesome work! I was wondering if we could use this method to have a powerful singing model as well ? Either text-to-singing or direct voice conversion ? Thanks
SWivid created a comment on an issue on SWivid/F5-TTS
closing this issue, feel free to open if further questions.
SWivid closed an issue on SWivid/F5-TTS
Segmentation fault (core dumped)
(f5tts) haoen@showlab04:~/F5-TTS$ python gradio_app.py Using cuda device Download Vocos from huggingface charactr/vocos-mel-24khz vocab : Emilia_ZH_EN tokenizer : pinyin model : /home/hao...
SWivid created a comment on a pull request on SWivid/F5-TTS
@lpscr Hi, we are having a general convertion to a package-compatible form of repo. Maybe we could do local develop temporarily and just wait for above-mentioned key update (in order to have less ...
PhamDangNguyen opened an issue on SWivid/F5-TTS
How can I use the fine-tune code for Vietnamese?
I wan to using code to training model for Vietnamese domains. I using phoneme for vocab. How I can get it?
SWivid created a comment on an issue on SWivid/F5-TTS
Yes, customed tokenizer is needed to match for changes.