Commit Graph

Select branches

Hide Pull Requests

dev

Mono Color

• 19d9f58009 create dummy model on the fly Eren Gölge 2021-04-27 13:27:24 +02:00
• 4719414f2e remove imports Eren Gölge 2021-04-27 11:25:17 +02:00
• add97cddc1 move function and remove import Eren Gölge 2021-04-27 11:22:56 +02:00
• 8f0519d203 bump up numpy version Eren Gölge 2021-04-27 11:13:57 +02:00
• 734e6a515c bug fix Eren Gölge 2021-04-27 10:27:45 +02:00
• 6bdd81667e place holders for sc-glow and hifigan models Eren Gölge 2021-04-26 19:49:18 +02:00
• 2f0716073e enable multi-speaker CoquiTTS models for synthesize.py Eren Gölge 2021-04-26 19:36:53 +02:00
• b531fa699c remove conflicy noise Eren Gölge 2021-04-26 15:27:52 +02:00
• f37b488876 Merge branch 'speaker-manager' of https://github.com/coqui-ai/TTS into speaker-manager Eren Gölge 2021-04-26 15:25:25 +02:00
• b82daa5e86 style and linter fixes Eren Gölge 2021-04-26 15:22:24 +02:00
• 20e42a3381 add save audio option Edresson 2021-04-23 15:00:00 -03:00
• 8228091f92 add script for extraction of tts spectrograms Edresson 2021-04-23 14:17:46 -03:00
• 4cf211348d styling and linting Eren Gölge 2021-04-23 17:57:27 +02:00
• a878d8fb42 update tests Eren Gölge 2021-04-23 17:55:02 +02:00
• 7eb0c60d2e let synthesizer to pass speaker encoder file paths to speaker manager Eren Gölge 2021-04-23 17:48:08 +02:00
• f69195739e let speaker manager compute mean x_vector from multiple wav files Eren Gölge 2021-04-23 17:46:50 +02:00
• 179722e3a7 new arguments to synthesize.py for loading speaker encoder and speaker wavs Eren Gölge 2021-04-23 17:46:21 +02:00
• dfa415a8b8 small refactor in server.py Eren Gölge 2021-04-23 15:53:23 +02:00
• c80d21f311 load speaker_encoder_ap and compute x_vector directly from the input file in speaker manager Eren Gölge 2021-04-23 15:52:34 +02:00
• ad047c8195 html formatting, enable multi-speaker model on the server with a dropdown menu to select the speaker Eren Gölge 2021-04-22 15:22:36 +02:00
• f9f3d04d14 remove moved function Eren Gölge 2021-04-22 12:37:12 +02:00
• 10c988ac8c update server.py Eren Gölge 2021-04-22 12:36:46 +02:00
• 32e6afc009 add SpeakerManager tests Eren Gölge 2021-04-21 13:28:43 +02:00
• 6d0f5e0459 use SpeakerManager in Synthesizer Eren Gölge 2021-04-21 13:13:27 +02:00
• e97126314c add ``unique`` argument to make_symbols to fix the incompat. issue of the SC-Glow models Eren Gölge 2021-04-21 13:12:35 +02:00
• d08888e603 formating speakers.py Eren Gölge 2021-04-21 13:12:11 +02:00
• df422223a3 initial SpeakerManager implementation Eren Gölge 2021-04-21 13:11:46 +02:00
• 7a7aeb35f5 fix the glow-tts in setup_model Eren Gölge 2021-04-21 13:10:40 +02:00
• d42748082a update argument name external_speaker_embedding_dim -> speaker_embedding_dim add inference_noise_scale argument to glow-tts Eren Gölge 2021-04-21 13:09:44 +02:00
• 2da81f5bb6 add load_chekpoint to speaker encoder Eren Gölge 2021-04-21 13:09:04 +02:00
• 1229ccbf07 update argument name in server.py Eren Gölge 2021-04-21 13:08:45 +02:00
• af2d36faeb update synthesize.py for multi-speaker setting Eren Gölge 2021-04-21 13:08:25 +02:00
• 99dc07a7dd add ``unique`` param to keep scglow models compatible (they are duplicate symbols ins the character set) Eren Gölge 2021-04-16 18:29:52 +02:00
• c955a12428 set the default layer size compatible with scglow Eren Gölge 2021-04-16 18:29:35 +02:00
• 3ace2440fa fix a mistake from rebase Eren Gölge 2021-04-16 19:39:47 +02:00
• aadb2106ec code styling Eren Gölge 2021-04-16 16:01:40 +02:00
• af7baa3387 refactoring to allow defining the speaker file externally Eren Gölge 2021-04-16 15:45:16 +02:00
• f393c08ad4 add usage of new Synthetizer class in the chinese model notebook kirianguiller 2021-03-01 15:18:40 +01:00
• 7dccbfdcd5 handle multi speaker and gst in Synthetizer class kirianguiller 2021-03-01 15:17:15 +01:00
• cc4efb437b Merge pull request #446 from WeberJulian/fix-windows Eren Gölge 2021-04-23 18:04:13 +02:00
• d2b6326b8b change optimizer initialization for compatibility with Hifi-GAN official implementation Edresson 2021-04-23 07:54:39 -03:00
• 4205284f92 Change name of the functions WeberJulian 2021-04-23 10:09:55 +02:00
• a26498181b Change back the default value WeberJulian 2021-04-22 16:10:17 +02:00
• 355e1f47ab fix dumb mistake Julian Weber 2021-04-22 15:50:29 +02:00
• c125b71f36 fix windows support Julian Weber 2021-04-22 15:14:24 +02:00
• a6cd04456e Merge branch 'dev' of https://github.com/coqui-ai/TTS into dev Eren Gölge 2021-04-22 12:39:12 +02:00
• f5fd7f78d4 server: also listen to ipv6 Jörg Thalheim 2021-04-16 09:26:36 +02:00
• ef37633cb3 [ci skip] use prenet_dropout by default with Tacotron models Eren Gölge 2021-04-22 12:37:31 +02:00
• 0ee3eeefb5 [ci skip] update CONTRIBUTING.md Eren Gölge 2021-04-21 13:55:50 +02:00
• 39ceb3f7df Update README.md Eren Gölge 2021-04-21 13:50:35 +02:00
• 757dfb9289 add SpeakerManager tests Eren Gölge 2021-04-21 13:28:43 +02:00
• e1d960da9e use SpeakerManager in Synthesizer Eren Gölge 2021-04-21 13:13:27 +02:00
• 04b6881b66 add ``unique`` argument to make_symbols to fix the incompat. issue of the SC-Glow models Eren Gölge 2021-04-21 13:12:35 +02:00
• 790946faec formating speakers.py Eren Gölge 2021-04-21 13:12:11 +02:00
• ab313814de initial SpeakerManager implementation Eren Gölge 2021-04-21 13:11:46 +02:00
• 09890c7421 fix the glow-tts in setup_model Eren Gölge 2021-04-21 13:10:40 +02:00
• 8764d02eb2 update argument name external_speaker_embedding_dim -> speaker_embedding_dim add inference_noise_scale argument to glow-tts Eren Gölge 2021-04-21 13:09:44 +02:00
• 8b40720977 add load_chekpoint to speaker encoder Eren Gölge 2021-04-21 13:09:04 +02:00
• 37cad38c27 update argument name in server.py Eren Gölge 2021-04-21 13:08:45 +02:00
• 9bccee9da8 update synthesize.py for multi-speaker setting Eren Gölge 2021-04-21 13:08:25 +02:00
• d0786be2a3 remove matrix link Eren Gölge 2021-04-19 11:07:40 +02:00
• e31f174cf8 Merge pull request #435 from Mic92/ipv6 Eren Gölge 2021-04-16 20:34:49 +02:00
• d2fa8add1f add ``unique`` param to keep scglow models compatible (they are duplicate symbols ins the character set) Eren Gölge 2021-04-16 18:29:52 +02:00
• d9612a4351 set the default layer size compatible with scglow Eren Gölge 2021-04-16 18:29:35 +02:00
• 1038fd420d fix a mistake from rebase Eren Gölge 2021-04-16 19:39:47 +02:00
• 47e356cb48 code styling Eren Gölge 2021-04-16 16:01:40 +02:00
• 25328aad00 refactoring to allow defining the speaker file externally Eren Gölge 2021-04-16 15:45:16 +02:00
• 83aa415934 add usage of new Synthetizer class in the chinese model notebook kirianguiller 2021-03-01 15:18:40 +01:00
• 48ae52a9a3 handle multi speaker and gst in Synthetizer class kirianguiller 2021-03-01 15:17:15 +01:00
• d0d7eae39a Merge pull request #430 from AXKuhta/main Eren Gölge 2021-04-16 13:10:47 +02:00
• 97f98e4c45 server: also listen to ipv6 Jörg Thalheim 2021-04-16 09:26:36 +02:00
• a53958ae3a fix urls for the new models Eren Gölge 2021-04-15 17:05:00 +02:00
• d63e4f288d Merge branch 'dev' into main Eren Gölge 2021-04-15 16:38:18 +02:00
• 4450c9bf5f [ci skip]update version Eren Gölge 2021-04-15 16:37:37 +02:00
• 9cc17be53a formatting and a small bug fix in Tacotron model Eren Gölge 2021-04-15 16:36:51 +02:00
• 1ad838bc83 add newly released models under .model.json Eren Gölge 2021-04-15 16:06:10 +02:00
• 7cada1a949 remove noise Eren Gölge 2021-04-15 15:30:45 +02:00
• d60a8d7211 show the real waveform on TB too for GAN vocoder training. Eren Gölge 2021-04-15 15:30:06 +02:00
• 5fbe926429 change the default TTS model to TacotronDDC Eren Gölge 2021-04-15 15:29:44 +02:00
• 993f4aed86 This snippet was trying to load the model as the config file AXKuhta 2021-04-15 11:53:27 +08:00
• bae2ad0ff5 add HiFiGAN to readme Eren Gölge 2021-04-15 03:15:12 +02:00
• 3de5a89154 optionally enable prenet dropout at inference time for tacotron models Eren Gölge 2021-04-13 13:24:56 +02:00
• c00212b67c bump up pytorch version Eren Gölge 2021-04-13 12:41:26 +02:00
• 28a2fed8a3 update hifigan in .model.json Eren Gölge 2021-04-12 16:48:05 +02:00
• abaf36861a aligntts model .model.json placeholder Eren Gölge 2021-04-12 16:43:52 +02:00
• 480e2f7888 docstring update and better handling make_symbols Eren Gölge 2021-04-12 16:40:49 +02:00
• b735076bb4 linter fixes Eren Gölge 2021-04-12 13:14:11 +02:00
• b11d1cb845 small fixes Eren Gölge 2021-04-12 12:40:55 +02:00
• a7f6045644 Merge branch 'reformat' into hifigan-reformat Eren Gölge 2021-04-12 12:00:17 +02:00
• f519012dea reformatting and styling Eren Gölge 2021-04-12 11:47:39 +02:00
• 9011dddf77 tacotron DDC placeholder in models.json Eren Gölge 2021-04-12 04:06:27 +02:00
• 3c0d1d066d update permissions of all the shell scripts Eren Gölge 2021-04-10 20:04:27 +02:00
• a750c07717 change file permissions to +x Eren Gölge 2021-04-10 19:51:48 +02:00
• d295d5de97 remove torch.no_grad from TorchSTFT Eren Gölge 2021-04-10 19:43:57 +02:00
• 5b70da2e3f restore schedulers only if training is continuing a previous training inherit nn.Module for TorchSTFT Eren Gölge 2021-04-09 19:31:28 +02:00
• 2c71c6d8cd [ci skip]update gan vocoder configs to reflect the recent changes Eren Gölge 2021-04-09 17:15:05 +02:00
• 8f2a4404b2 [ci skip] lang. update for CONTRIBUTING.md Eren Gölge 2021-04-09 17:07:37 +02:00
• 72e72a9476 adapt Makefile for CI Eren Gölge 2021-04-09 16:52:24 +02:00
• 4e47d8f618 Update the main README.md Eren Gölge 2021-04-09 13:51:29 +02:00
• 1cdf8498d7 new CONTRIBUTING.md Eren Gölge 2021-04-09 13:51:04 +02:00