Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
mhenrichsen created a comment on an issue on SWivid/F5-TTS
@SWivid no. It's 600ish hours of 500k samples. Nothing longer than 25 seconds.
SWivid created a comment on an issue on SWivid/F5-TTS
How is the dataset constructed. Does it contain extremely few long samples?
mhenrichsen opened an issue on SWivid/F5-TTS
VRAM usage randomly spikes. Continues fine after restart.
Hi, I'm training on a single A100. I've tried multiple batch sizes, but the same thing keeps happening. I usually use between 10-25% of all available GPU VRAM, and then all of the sudden it ...
danielw97 opened an issue on SWivid/F5-TTS
Ability to specify finetuned model in the Gradio app
Hello, I've come to this repo recently, so apologies if I've missed something here. Things sound great, although I'm currently doing some finetuning tests from the original f5tts checkpoint with ...
SWivid created a comment on an issue on SWivid/F5-TTS
> `<Speaker1>` instead of (Speaker1)? or `|<Speaker1>|`
chigkim created a comment on an issue on SWivid/F5-TTS
Oops, Github stripped the characters. lol Maybe, `<Speaker-1>` instead of (Speaker1)?
chigkim created a comment on an issue on SWivid/F5-TTS
Also, I'm thinking it should use characters less likely to encounter in the main text to markup voices? Maybe, <Speaker-1> instead of (Speaker1)?
andriken created a comment on an issue on SWivid/F5-TTS
> So far, the finetune I'm running has seen 1700 hours of Japanese data (1 epoch) and it's starting to learn how to read out sentences correctly in Japanese. Given this, I'm gonna see how it sounds...
ChibiChubu opened an issue on SWivid/F5-TTS
i got these error when train data
accelerate launch finetune-cli.py --exp_name F5TTS_Base --learning_rate 1e-05 --batch_size_per_gpu 3200 --batch_size_type frame --max_samples 6 --grad_accumulation_steps 1 --max_grad_norm 1 --epoch...
SWivid created a comment on an issue on SWivid/F5-TTS
Yes, you could just remove it. thought ``` Multi-Style (Speaker1_Happy) Hello (Speaker2_Sad) Hi (Speaker1_Main) What's going on? ``` is great, with newline is more clear
SWivid created a comment on an issue on SWivid/F5-TTS
可以 用`inference-cli.py` 在手动输入参考音频对应的文本（即不使用whisper转录文本）的情况下只要2G不到 也可以修改`gradio_app.py` prune掉不需要的部分
chigkim opened an issue on SWivid/F5-TTS
Podcast vs Multi-Style Tabs on Gradio App?
@SWivid @jpgallegoar @JarodMica @lpscr @cocktailpeanut, are there any difference in multi-style and podcast tabs other than markup format? ###Podcast Speaker1: Hello Speaker2: Hi ... ### Mul...
