Comments (5)
The warning you got is expected and your error doesn't seem related! I've never encountered it, did you double-check your dataset? Feel free to send details!
from parler-tts.
you can remove that message by modifying a part of the DataCollatorEncodecWithPadding in the training script.
@dataclass
class DataCollatorEncodecWithPadding:
"""
Data collator that will dynamically pad the inputs received to the longest sequence in the batch or
to `max_length` if `max_length` is set and `padding=max_length`.
"""
feature_extractor: AutoFeatureExtractor
audio_column_name: str
feature_extractor_input_name: Optional[str] = "input_values"
max_length: Optional[int] = None
padding: Optional[str] = "longest"
def __call__(self, features: List[Dict[str, Union[List[int], torch.Tensor]]]) -> Dict[str, torch.Tensor]:
# split inputs and labels since they have to be of different lengths and need
# different padding methods
audios = [feature[self.audio_column_name]["array"] for feature in features]
len_audio = [len(audio) for audio in audios]
# here
sampling_rate = self.feature_extractor.sampling_rate
batch = self.feature_extractor(
audios, sampling_rate=sampling_rate, return_tensors="pt", padding=self.padding, max_length=self.max_length
)
batch["len_audio"] = torch.tensor(len_audio).unsqueeze(1)
return batch
from parler-tts.
Hey @choihk6610, thanks!
feel free to open a PR to have this! I can do it if necessary
from parler-tts.
@ylacombe thank you for your response.
i have just opened a simple pr and would appreciate it if you could review it.
from parler-tts.
It has been merged, thanks for your help!
from parler-tts.
Related Issues (20)
- Error on fine tuning HOT 9
- both small and large checkpoint mispronounces some words
- Request for prompt guide to write description
- Impact of RoPE and Adding Prompts in Cross Attention on Model Performance
- Torch compile problem and some more ideas HOT 3
- How many epoch of training did you do? What is the accuracy?
- asking for gpu to finetune large model HOT 1
- Speaker voice is not consistent across different generation HOT 2
- Stream and play HOT 2
- Hope to support Chinese.
- How to use italian language?
- Documentation references to facebook/parler_tts-small
- Substitute different Audio codec and Text encoder?
- ValueError: 'dac' is already used by a Transformers config, pick another name. HOT 1
- [Performance] Usage with `optimum-quanto`? HOT 2
- Using static cache and torch.compile, generating more tokens than previously during warmup will trigger graph recapture HOT 5
- any list of all 36 voices? HOT 5
- Alltalk Integration & 1x question re voice consistency HOT 2
- some question to prepare multilinguality training from scratch HOT 2
- GGML implementation when? HOT 6
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from parler-tts.