kunibald

how to read form this pipe? write to file or whatever https://audiopipe.suno.ai/?item_id=

Which speaker id corresponds to which vctk speaker, is this info still available? Interesting for resuming training.

hi, what changes would need to be made in order to train a pretrain model with hubert soft (https://github.com/bshall/soft-vc)?

sls | Return FAILURE status to client when internal `_handle_result` fails

3

**Is your feature request related to a problem? Please describe.** Client gets response with status "COMPLETED" without any response data that the handler is adding. This message is then observed...

2nd-order Heun's method for ODE sampling

2

as mentioned here https://openreview.net/forum?id=zNA7u7wtIN&noteId=aiIagkNqgq https://openreview.net/attachment?id=Ymg8GZvGds&name=pdf you have a take on this and would it be easy to implement and try?

Multi gpu training; RuntimeError: [...] LightningModule has parameters that were not used in producing the loss returned by training_step.

4

added `devices="auto"` in `train.py` to utilize multiple gpus ``` trainer: Trainer = hydra.utils.instantiate(cfg.trainer, callbacks=callbacks, logger=logger, devices="auto") ``` Training terminates shortly after start with this error: ``` RuntimeError: It looks like...

TypeError: pflowTTS.synthesise() got an unexpected keyword argument 'spks'

2

to infer on custom model (single speaker), this is called: `python pflow/cli.py --file $INPUT_FILE --checkpoint_path "$CHECKPOINT_PATH" --vocoder "hifigan_univ_v1" --spk $SPEAKERID` it loads custom model and vocoder ``` [!] Loading custom_model!...

possible custom model inference issue

1

``` if not hasattr(args, "checkpoint_path") and args.checkpoint_path is None: model_path = args.checkpoint_path ``` shouldn't this be a positive check? ``` if hasattr(args, "checkpoint_path") and args.checkpoint_path: model_path = args.checkpoint_path ``` https://github.com/p0p4k/pflowtts_pytorch/blob/82c70bf71e082356c785fcdd8fc24eea912abfc4/pflow/cli.py#L68C17-L68C17

Hi, I have to say that the extensive and thorough documentation is much appreciated! Would the mediapipe mocap presented in v4 demo https://github.com/pkhungurn/talking-head-anime-4-demo also work for v3? Thank you for...

shouldn't it be set to True? https://github.com/runpod/runpod-python/blob/f6971a8f142a7b4796fb0f1ffa3b3f201d130a81/runpod/serverless/modules/rp_http.py#L77C16-L77C17