Welcome to mirror list, hosted at ThFree Co, Russian Federation.

gitlab.xiph.org/xiph/opus.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/dnn/torch
AgeCommit message (Expand)Author
2023-12-20Merge LACE/NoLACE under OSCE frameworkopus-ng-lace-integration5Jan Buethe
2023-11-16Smaller version of farganJean-Marc Valin
2023-11-15Using sparse GRUs in DRED decoderJean-Marc Valin
2023-11-09Dumping RDOVAE stats from XMLJean-Marc Valin
2023-11-08DRED: quantize scale and dead zone to 8 bitsJean-Marc Valin
2023-11-07added weight export script for LACE/NoLACEJan Buethe
2023-11-07added transposed 1d convolutions to wexchangeJan Buethe
2023-11-07Split stats in two and remove useless dimensionsJean-Marc Valin
2023-11-06updated moc to match results in ietf118 presentationJan Buethe
2023-11-06DRED: quantize r and p0 parameters with 8 bitsJean-Marc Valin
2023-11-02updated moc methodJan Buethe
2023-10-29Pitch and fargan model updatesJean-Marc Valin
2023-10-20Quantizing pitchdnn and rdovae weightsJean-Marc Valin
2023-10-20added 16kHz version of opus_compare in pythonJan Buethe
2023-10-20finalized quantization option in export_rdovae_weights.pyJan Buethe
2023-10-19Marking RDOVAE layers to quantizeJean-Marc Valin
2023-10-19prepared quantization implementation for DREDJan Buethe
2023-10-19restructured osce readmeJan Buethe
2023-10-19added LACE/NoLACE checkpoint URLJan Buethe
2023-10-14C implementation of FARGANJean-Marc Valin
2023-10-10update fargan to match version 45Jean-Marc Valin
2023-10-07updated osce readmeJan Buethe
2023-10-06Switching to neural pitch estimatorJean-Marc Valin
2023-10-02Using a DenseNet for DREDJean-Marc Valin
2023-10-02Remove unneeded (I think) tanh at the endJean-Marc Valin
2023-10-01First version of pitch DNN C codeJean-Marc Valin
2023-09-30Fix model savingJean-Marc Valin
2023-09-29updated PitchDNN export scriptJan Buethe
2023-09-29refactoring and cleanupJan Buethe
2023-09-29fixed type in error messageJan Buethe
2023-09-29added rudimentary support for dumping nn.Conv2d layersJan Buethe
2023-09-27No features skip needed to align pitch featuresJean-Marc Valin
2023-09-27Script to compute the groundtruth data using CREPEJean-Marc Valin
2023-09-26updated default values in adv_train_fargan.pyJan Buethe
2023-09-26changed checkpoint formatJan Buethe
2023-09-26Adapting to new data format/modelJean-Marc Valin
2023-09-26Python code for neural pitchKrishna Subramani
2023-09-26RDOVAE model updateJean-Marc Valin
2023-09-22bugfixJan Buethe
2023-09-22pulse-related bugfixJan Buethe
2023-09-21more lavoce stuffJan Buethe
2023-09-21Quantizing initial state with rdovae tooJean-Marc Valin
2023-09-20added FARGAN adversarial training scriptJan Buethe
2023-09-13removed trailing whitespace in farganJan Buethe
2023-09-13integrated JM's FFT ada convJan Buethe
2023-09-13Remove --has-gain and --passthrough-size argsJean-Marc Valin
2023-09-13Add conditioning interpolation, fwconv layerJean-Marc Valin
2023-09-13Only use one frame of pre-loadingJean-Marc Valin
2023-09-13Randomly double the training sequence lengthJean-Marc Valin
2023-09-13Adds skip connectionsJean-Marc Valin
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-28 18:39:49 +0300