{"id":27373372,"url":"https://github.com/nipponjo/mixer-tts-pytorch","last_synced_at":"2026-05-01T12:31:53.502Z","repository":{"id":287557270,"uuid":"965064801","full_name":"nipponjo/mixer-tts-pytorch","owner":"nipponjo","description":"Mixer-TTS for efficient TTS","archived":false,"fork":false,"pushed_at":"2025-04-12T15:11:58.000Z","size":1273,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-10-11T04:44:15.731Z","etag":null,"topics":["deep-learning","gan","ljspeech","mixer-tts","python","pytorch","speech","speech-synthesis","text-to-speech","torchaudio","tts","tts-model","voice-synthesis"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nipponjo.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-04-12T10:33:25.000Z","updated_at":"2025-08-31T15:53:12.000Z","dependencies_parsed_at":"2025-04-13T11:14:32.906Z","dependency_job_id":null,"html_url":"https://github.com/nipponjo/mixer-tts-pytorch","commit_stats":null,"previous_names":["nipponjo/mixer-tts-pytorch"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/nipponjo/mixer-tts-pytorch","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Fmixer-tts-pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Fmixer-tts-pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Fmixer-tts-pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Fmixer-tts-pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nipponjo","download_url":"https://codeload.github.com/nipponjo/mixer-tts-pytorch/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Fmixer-tts-pytorch/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32497810,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-30T13:12:12.517Z","status":"online","status_checked_at":"2026-05-01T02:00:05.856Z","response_time":64,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","gan","ljspeech","mixer-tts","python","pytorch","speech","speech-synthesis","text-to-speech","torchaudio","tts","tts-model","voice-synthesis"],"created_at":"2025-04-13T11:14:31.557Z","updated_at":"2026-05-01T12:31:53.475Z","avatar_url":"https://github.com/nipponjo.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# mixer-tts-pytorch\n\n[[Samples]](https://nipponjo.github.io/tts-mixer-samples/)\n\nThis repo contains an implementation of the Mixer-TTS model ([https://arxiv.org/abs/2110.03584](https://arxiv.org/abs/2110.03584)).\n\nPre-trained weights are available for the [LJ Speech](https://keithito.com/LJ-Speech-Dataset/) dataset. \nThe channel dimensions of the convolutions inside the model were chosen as 384, 128 and 80, resulting in models with 20.6M, 3.17M and 1.74M parameters.\n\nThe pre-trained models take IPA symbols as input. Please refer to [here](https://bootphon.github.io/phonemizer/install.html) to install `phonemizer` and the `espeak-ng` backend.\n\nA simple patch-based discriminator was used in training to generate more natural mel spectrograms.\n\nAudio samples are available [here](https://nipponjo.github.io/tts-mixer-samples/).\n\n**Pre-trained models**\n\nAll (3) checkpoint files can be downloaded by running: `python download_files.py`\n\n|Dataset|dim|params|name|link|\n|-------|---|------|-----|---|\n|LJSpeech|80|1.74M|mixer_lj_80|[link](https://drive.google.com/file/d/1YTiA6S3okiuX-_AttUhJNVgiPzVYAyjv/view?usp=sharing)|\n|LJSpeech|128|3.17M|mixer_lj_128|[link](https://drive.google.com/file/d/1wVvOyaBLxqrKAssXmEYG9mszZsqEaX5R/view?usp=sharing)|\n|LJSpeech|384|20.6M|mixer_lj_384|[link](https://drive.google.com/file/d/16Rq99ZmXVfiDE_nsxmUBzF3XKEOUh5wx/view?usp=sharing)|\n\n\nThe pre-trained models output the 80-channel mel spectrogram version first proposed by the [HiFi-GAN](https://github.com/jik876/hifi-gan) vocoder.\n\n\n**References**\n\nThe model was taken out of NVIDIA's [NeMo](https://github.com/NVIDIA/NeMo) framework in order to make it easier to modify and have fewer dependencies. An energy embedding and optional speaker and emotion embeddings have been added.\n\n[Mixer-TTS in NeMo](https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/tts/models/mixer_tts.py)\n\nPaper:\n```\n@article{Tatanov2021MixerTTSNF,\n  title={Mixer-TTS: Non-Autoregressive, Fast and Compact Text-to-Speech Model Conditioned on Language Model Embeddings},\n  author={Oktai Tatanov and Stanislav Beliaev and Boris Ginsburg},\n  journal={ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  year={2021},\n  pages={7482-7486},\n}\n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnipponjo%2Fmixer-tts-pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnipponjo%2Fmixer-tts-pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnipponjo%2Fmixer-tts-pytorch/lists"}