2023-05-01T21:08:19Z - 2024-05-01T21:08:19Z
Overview
9 Pull requests merged by 3 users
Merged
#393 Freeze beartype==0.15.0
Merged
#369 master
Merged
#350 Websocket fixes / additions
Merged
#341 fix filename generation which didn't work and overwrote existing files
Merged
#336 favor existing arguments from parameters (kwargs) over global (args)
Merged
#334 websocket server: API change(!), better response format
Merged
#333 websocket server: small fix
Merged
#328 added simple websocket server which allows to start tts generation tasks, retrieving autoregressive models and voices list
Merged
#301 Freeze pydantic package to 1.10.11
4 Pull requests proposed by 3 users
Proposed
#448 Added a CPU setup script (based on the existing CUDA and ROCm setup scripts)
Proposed
#455 Add Hifigan compatibility to this repo
Proposed
#474 add whisper large-v3
Proposed
#475 allow using 0.01 in save frequncy for large datasets
66 Issues closed from 44 users
Closed
#462 Tortoise seems to no longer be included in the install. Just errors. Same with DLAS.
Closed
#460 Not found torch_python.dll or one of its dependencies & 'utf-8' codec can't decode byte 0x84 in position 0: invalid start byte
Closed
#452 Error compile with TORCH_USE_CUDA_DSA
Closed
#415 RESTful API?
Closed
#427 Google Colab Notebook Not Working
Closed
#429 Resuming Training keyError: 'iter'
Closed
#426 Kickstart foreign language training using XTTS weights?
Closed
#424 RuntimeError: Error building extension 'transformer_inference'
Closed
#416 Can't get the model training started
Closed
#406 selected index k out of range when attempting to gen more than 2 candidates
Closed
#409 Fresh install, no module named dlas
Closed
#402 Very Bad Training Time & Results.
Closed
#398 Issue #152 Inaccessable
Closed
#396 Validation.txt shows nothing
Closed
#397 web interface
Closed
#384 Why so many models? And about a thousand of other questions :)
Closed
#395 Why does my AMD GPU eat up too much vram?
Closed
#390 Is Training Console Output Broken?
Closed
#389 Recreate dataset after correcting whisper.json doesn't overwrite the train.txt
Closed
#382 Blocked while training attempt - name 'str2optimizer8bit_blockwise' is not defined"
Closed
#380 Error no kernel image is available for execution on the device at line 167
Closed
#377 Attempting to run training -- libcudart.so not t found
Closed
#376 Unable to find complete code
Closed
#368 argument of type 'NoneType' is not iterable
Closed
#367 NameError: name 'VOCOS_ENABLED' is not defined when using bark
Closed
#366 Extremely large output file size using tortoise
Closed
#364 Missing dataset: ./training/voice1//whisper.json
Closed
#352 Best ways to get rid of static?
Closed
#353 Resume training with expanded dataset
Closed
#345 [Dataset] Where should I put the dataset and what should it look like?
Closed
#348 Is it possible to train faster tortoise tts with this?
Closed
#347 ModuleNotFoundError: No module named 'tortoise'
Closed
#324 Does bark feature fine tuning?
Closed
#343 Vall-E Backend Training: RuntimeError: Failed to find any .qnt.pt file in specified training dataset.
Closed
#335 (Prepare Dataset) List indices must be intergers or slices, not dict
Closed
#339 Vall-E Backend Training: "list indices must be integers or slices, not dict"
Closed
#340 Vall-E Backend: AttributeError: 'Config' object has no attribute 'hdf5'
Closed
#332 Error when training
Closed
#326 WhisperX models (Large issue)
Closed
#329 Dataset too large?
Closed
#327 Training bugs
Closed
#320 Error when running start.bat after installing.
Closed
#315 Is it possible to introduce voice from file instead of mic?
Closed
#312 Browser crashes probably shouldn't pause training
Closed
#267 Are conditioning latents harder to generate for larger datasets?
Closed
#310 Cannot get the cli to work
Closed
#309 Tortoise->normalize->rvc
Closed
#299 Error when changing any settings in webui
Closed
#298 chunk(): argument 'chunks' (position 2) must be int, not NoneType
Closed
#297 Google Text-to-Speech
Closed
#294 no module named "tortoise"
Closed
#293 Suggestion on how to package as Docker container
Closed
#284 Unable to complete training.
Closed
#279 Missing dataset: whisper.json.
Closed
#276 [Training] Detected call of `lr_scheduler.step()` before `optimizer.step()`
Closed
#271 Output wav filesizes are huge on a trained model
Closed
#263 cli.py loads model everytime it's ran
Closed
#234 Graphs follow the wrong number (steps instead of epochs)
Closed
#226 Weird groaning at the end of every output file...
Closed
#128 Has anyone managed to train a voice to be able to shout?
Closed
#258 A TorToise TTS model fine-tuned to speak in Russian
Closed
#254 No module named 'dlas'
Closed
#244 Step by step data prep and training/finetuning guide?
Closed
#246 Is it possible to load a different Autoregressive Model through the Gradio API?
Closed
#232 ModuleNotFoundError: No module named 'dlas' when trying to run Training
Closed
#221 Getting total gibberish when finetuning on a new language
183 Issues created by 134 users
Opened
#227 Error when setting CVVP value.
Opened
#228 Bug/Issue: Incorrect save steps passed to train.yaml
Opened
#229 TypeError: Progress.tqdm() got an unexpected keyword argument 'track_tqdm'
Opened
#230 tqdm.write() got an unexpected keyword argument 'desc'
Opened
#231 is there a better way to generate longer passages of text?
Opened
#233 Thanks for all the voice synthesis threads
Opened
#235 Graph not updating after a day
Opened
#236 Phenomizer expects a list of strs, not one giant str
Opened
#237 Tortoise Training: 'num_conditioning_inputs' is useless
Opened
#238 Another share your models thread.
Opened
#239 IndexError: list index out of range
Opened
#240 Gradio missing attributes on fresh install
Opened
#241 ImportError: cannot import name 'get_voice_dir' from 'tortoise.utils.audio'
Opened
#242 Is it possible to generate using the command-line?
Opened
#243 Output sounds slow and lower pitch to tortoise-tts
Opened
#245 Transcript Generation
Opened
#247 Unable to start the app using start.bat
Opened
#248 Overfitting with large datasets
Opened
#249 Out of memory errors and using whisperX
Opened
#250 resume a training
Opened
#251 "Iterations" when generating
Opened
#252 Whats changed recently? finetuning in webUI is completely broken now
Opened
#253 Results, Retrospectives, and Recommendations
Opened
#255 Japanese Tokenizer: Error while attempting to unpickle Tokenizer
Opened
#256 ModuleNotFoundError: No module named 'tortoise'
Opened
#257 Anyone tried Muti-lingual for Tortoise ?
Opened
#259 A Tortoise TTS Model Fine-Tuned to Speak in Russian
Opened
#260 ImportError: DLL load failed while importing torch_directml_native: The specified process was not found
Opened
#261 Question; How to continually run from CLI
Opened
#262 [Tool] for reviewing / trimming whisperX generated audio
Opened
#264 Is there anyway to save the voice from a random generation ?
Opened
#265 Whisper transcribing isn't doing all the files
Opened
#266 "Original Latents Method" - What is this for, and what is the best option?
Opened
#268 Install halts
Opened
#269 Custom tokenizer
Opened
#270 models merging
Opened
#272 Training issue
Opened
#273 Maybe dumb question; Could running multiple instances share the same models/resources?
Opened
#274 Can't use whisperX.
Opened
#275 [Feature Req] Allow deletion of bad/culled files and using other sample rates for data splitting.
Opened
#277 Google Translatotron 3: Speech to Speech Translation with Monolingual Data
Opened
#278 I ran the update.bat and now it won't run (screen shot attached)
Opened
#280 Error when running start.bat
Opened
#281 Attempting to restart training doesn't actually restart the training.
Opened
#282 Model sounds bad and nothing like the original
Opened
#283 Large dataset finetuning
Opened
#285 Training with Deutsche Audio Files, speaks english. How to make Deutsche Models?
Opened
#286 Deep voices
Opened
#287 Voice Chunk Size
Opened
#288 Did something break with the hugging face models ?
Opened
#289 conditioning_length: 44000 is different to sample rate?
Opened
#290 No module named 'tortoise.api'
Opened
#291 Error on generation: ValueError: Value too wide (>4 bytes) & other issues
Opened
#292 UnicodeDecodeError: 'utf-8' codec can't decode byte 0x84 in position 0: invalid start byte
Opened
#295 training
Opened
#296 Audio artifacts/repetitive words after training
Opened
#300 Voice never generates and gives error in console. windows 11
Opened
#302 Error with training can't find dlas module, tried manual install
Opened
#303 No module named 'rpds.rpds' I am not fully sure why
Opened
#304 training has bunch of errors and warnings not sure new system
Opened
#305 Need help in adopting your tortoise implementation
Opened
#306 "System error" when using Bark as TTS
Opened
#307 Training never produces usable voices
Opened
#308 Error: 'utf-8' codec can't decode byte 0xbd in position 0: invalid start byte
Opened
#311 How to train Russian languague? ERROR 'utf-8' codec can't decode byte 0x84 in position 0: invalid start byte
Opened
#313 Rust compiler is required on Windows
Opened
#314 How to provide Dynamic Prompt Setting Editing (switch between voices)
Opened
#316 getting error when using emotions
Opened
#317 Cant get Training running
Opened
#318 Thanks to MrQ and the Community and Question about Possible Donations.
Opened
#319 Transcription and diarization (speaker identification) - Easy dataset building?
Opened
#321 Error when installing Bark after having installed the main program
Opened
#322 Non-English Tokenizer
Opened
#323 change max duration of duration and text length on train yaml
Opened
#325 Weird sound after most lines
Opened
#330 White page on gradio.live when launched via colab notebook
Opened
#331 Have problem cloning voice
Opened
#337 Starting the vall-e backend crashes
Opened
#338 FileNotFoundError: [WinError 3] The system cannot find the path specified: '/home/user/aivoice/models/tortoise/autoregressive.pth'
Opened
#342 German Tokenizer
Opened
#344 ImportError: DLL load failed while importing torch_directml_native: The specified procedure could not be found.
Opened
#346 Japanese Tokenizer Issues
Opened
#349 unload_tts() doesnt unload the voice model from video memory
Opened
#351 AttributeError: 'NoneType' object has no attribute 'killed'
Opened
#354 Illegal Instruction after setting quality to anything above ultra-fast?
Opened
#355 RAM memory leak
Opened
#356 Loss gets stuck after resume training
Opened
#357 Finished training, but it hasn't
Opened
#358 The content of the generated sound is not correct
Opened
#359 Are YouTube rips entirely unusable for finetuning?
Opened
#360 The web page displays an error problem
Opened
#361 American imposter
Opened
#362 WhisperX installation
Opened
#363 Any tips for getting the fastest inference physically possible?
Opened
#365 Using TPUs in Google Colab?
Opened
#370 No Module named Deepspeed
Opened
#371 getting strange error when trying to train
Opened
#372 Errors when running start-cuda.bat
Opened
#373 RuntimeError: CUDA error: the launch timed out and was terminated CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
Opened
#374 Speed Increase?
Opened
#375 ImportError: DLL load failed while importing torch_directml_native: The specified procedure could not be found.
Opened
#378 How to train through Linux shell?
Opened
#379 Sharing a German fine-tuned model and Latin-1 tokenizer
Opened
#381 training model from scratch
Opened
#383 SSL: CERTIFICATE_VERIFY_FAILED
Opened
#385 mel loss and dataset questions
Opened
#386 Implementing XTTS by coqui?
Opened
#387 No module named 'tortoise' WSL WIndows 11
Opened
#388 beartype.roar.BeartypeDecorHintPep484Exception: Function torch.onnx.symbolic_helper._onnx_unsupported() return PEP 484 type hint "typing.NoReturn" invalid in this type hint context
Opened
#391 beartype conflict
Opened
#392 lr_scheduler
Opened
#394 ImportError: DLL load failed while importing torch_directml_native: The specified procedure could not be found.
Opened
#399 Nvidia Driver Woes - Super slow training
Opened
#400 Upscaled output creates bad quality?
Opened
#401 Elevens Labs Adam Premade voice T-TTS/RVC model weights
Opened
#403 Very Bad Training Time & Results.
Opened
#404 Deepspeed - Windows (Yes I know)
Opened
#405 Does not run in Docker
Opened
#407 OSError when running start.sh after fresh install
Opened
#408 UnicodeDecodeError: 'utf-8' codec can't decode byte 0x85 in position 16: invalid start byte
Opened
#410 Learning state restart question
Opened
#411 Cant get training to work? (windows 10 nvida GPU)
Opened
#412 Tortoise-TTS - Something went wrong. Connection errored out.
Opened
#413 `MemoryError` after doing a git pull
Opened
#414 How to export finetuned Tortoise model?
Opened
#417 Line delimiter is heard in the output
Opened
#418 Two trainings with exact same parameters result in different curves
Opened
#419 Double Counting Epochs?
Opened
#420 Chinese tortoise-tts wechat discuss group
Opened
#421 Finetuning diffusion model
Opened
#422 Training does not commence - missing module 'axial-positional-embedding'
Opened
#423 20 min audio, 500 epochs, underwhelming result
Opened
#425 CUDA out of memory error after installing and toggling on deepspeed
Opened
#428 Discussions Tab?
Opened
#430 American Accent training data gives British Accented results
Opened
#431 About updating to latest tortoise-tts repo
Opened
#432 XTTS trainer has been released
Opened
#433 Training multiple finetunes on a single dataset in GUI?
Opened
#434 <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1129)>
Opened
#435 [Discussion] What is a large dataset?
Opened
#436 Can't identify what's wrong (potentially list index out of range)
Opened
#437 How do I delete the app?
Opened
#438 Request: Implement StyleTTS2 by yl4579?
Opened
#439 New model to speed up and improve transcriptions
Opened
#440 How do i run the project on CPU?
Opened
#441 [Question] Learning Rate Schedule default?
Opened
#442 Vocoder finetuning
Opened
#443 problem with finetuning into an arabic language
Opened
#444 [Question] 124GB Model from 8 minutes of audio
Opened
#445 Can emotion be trained?
Opened
#446 [Feature Request] Add saving/loading of Generate configuration
Opened
#447 Issue with training in Colab
Opened
#449 XTTS-2 released
Opened
#450 [Q] What if I have train.txt but not whisper.json
Opened
#451 ai-voice-cloning install on Linux (Ubuntu) - 9642 Illegal Instruction Error
Opened
#453 Help Needed Problem Training
Opened
#454 StyleTTS2 - New Free Voice Cloning TTS
Opened
#456 Something went wrong Expecting value: line 1 column 1 (char 0)
Opened
#457 Can't start.bat
Opened
#458 ModuleNotFoundError: No module named 'torch'
Opened
#459 help installing tortoisee
Opened
#461 Processed Dateset and Missing Dataset Error
Opened
#463 Start.bat error after installation - "cannot import name 'broadcat'"
Opened
#464 torch.cuda.OutOfMemoryError: CUDA out of memory. OOM
Opened
#465 A very strange Issue: "Exception: Empty dataset Error"
Opened
#466 "Unsupported audio format provided: .pth"
Opened
#467 Training using "ipa" tokenizers
Opened
#468 [Question] Segments : what are their role in the training?
Opened
#469 What to do with models that come with a .index file?
Opened
#470 ModuleNotFoundError: No module named 'pyfastmp3decoder'
Opened
#471 Task exception was never retrieved - When generating or selecting any voice.
Opened
#472 Program no Longet Writes Train.txt File
Opened
#473 metavoicio / metavoice-1B-v0.1
Opened
#476 Cannot check for updates
Opened
#477 Colab: CUDA out of memory error before any voice is produced and when "Use CUDA for Voice Fixer" is turned off
Opened
#478 Noise at the end of generated voice
Opened
#479 [Sharing] Sharing my audio dataset manager
Opened
#480 Missing whisper and tensor error
Opened
#481 Slow loading of training data
Opened
#482 ModuleNotFoundError: No module named 'tortoise.utils.device'
Opened
#483 Getting error in start.sh script running after whisper detection
Opened
#484 So is this abandoned?
Opened
#485 ./start.sh: line 4: 30468 Segmentation fault (core dumped) python3 ./src/main.py "$@"
14 Unresolved Conversations
Open
#147
Discussion about Fine Tuning on a different language.
Open
#152
VALL-E Integration (and In Response To TorToiSe: a Quick Retrospective)
Open
#205
CUDA out of memory when generating a second clip
Open
#160
Can't train a single good model
Open
#166
utf-8 codec can't decod ebyte 0x81 in position 2
Open
#183
generating voice clip is so much slower compared to using original Tortoise TTS
Open
#121
Cuda OOM error when running start.bat
Open
#225
Requesting tips to make inference as fast as possible
Open
#60
Share your models
Open
#219
mrq adding significant American accent to same voice samples from tortoise-fast-tts
Open
#212
Error when training : TypeError: new(): invalid data type 'str'
Open
#222
Adding faster-whisper backend
Open
#223
Trying to Transcribe but getting error: 'Missing dataset: ./training/voice1//whisper.json'
Open
#224
All finetuned models are unstable when synthesizing lengthy content