{"id":30295792,"url":"https://github.com/nipponjo/tts_arabic","last_synced_at":"2025-12-30T09:27:37.361Z","repository":{"id":234814208,"uuid":"789538812","full_name":"nipponjo/tts_arabic","owner":"nipponjo","description":"TTS for Arabic (FastPitch, Mixer-TTS) in the ONNX format","archived":false,"fork":false,"pushed_at":"2025-07-19T12:11:06.000Z","size":88,"stargazers_count":25,"open_issues_count":2,"forks_count":4,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-07-19T16:54:20.412Z","etag":null,"topics":["arabic","arabic-tts","fastpitch","hifigan","mixer-tts","multi-speaker-tts","onnx","onnxruntime","python","speech","speech-synthesis","text-to-speech","tts","tts-model","vocos","voice-synthesis"],"latest_commit_sha":null,"homepage":"","language":"Python","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":"2024-04-20T20:44:08.000Z","updated_at":"2025-07-19T12:11:10.000Z","dependencies_parsed_at":"2024-05-28T21:46:33.701Z","dependency_job_id":"30a4a383-2266-457a-a3a7-a95d44158d98","html_url":"https://github.com/nipponjo/tts_arabic","commit_stats":null,"previous_names":["nipponjo/tts_arabic"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/nipponjo/tts_arabic","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Ftts_arabic","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Ftts_arabic/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Ftts_arabic/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Ftts_arabic/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nipponjo","download_url":"https://codeload.github.com/nipponjo/tts_arabic/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nipponjo%2Ftts_arabic/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":270796295,"owners_count":24647330,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-17T02:00:09.016Z","response_time":129,"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":["arabic","arabic-tts","fastpitch","hifigan","mixer-tts","multi-speaker-tts","onnx","onnxruntime","python","speech","speech-synthesis","text-to-speech","tts","tts-model","vocos","voice-synthesis"],"created_at":"2025-08-17T02:02:03.651Z","updated_at":"2025-12-30T09:27:37.356Z","avatar_url":"https://github.com/nipponjo.png","language":"Python","funding_links":[],"categories":["TTS Models"],"sub_categories":[],"readme":"Arabic TTS models (FastPitch, MixerTTS) from the [tts-arabic-pytorch](https://github.com/nipponjo/tts-arabic-pytorch) repo in the ONNX format — usable as a Python package for offline speech synthesis.\n\nAudio samples can be found [here](https://nipponjo.github.io/tts-arabic-speakers).\n\n**Install with**\n```\npip install git+https://github.com/nipponjo/tts_arabic.git\n```\n\n\n**Examples**\n```python\n\n# %%\nfrom tts_arabic import tts\n\n# %%\ntext = \"اَلسَّلامُ عَلَيكُم يَا صَدِيقِي.\"\nwave = tts(text, speaker=2, pace=0.9, play=True)\n\n# %% Buckwalter transliteration\ntext = \"\u003eals~alAmu Ealaykum yA Sadiyqiy.\"\nwave = tts(text, speaker=0, play=True)\n\n# %% Unvocalized input\ntext_unvoc = \"القهوة مشروب يعد من بذور البن المحمصة\"\nwave = tts(text_unvoc, play=True, vowelizer='shakkelha')\n\n```\n\n**Pretrained models**\n\n|Model|Model ID|Type|#params|Paper|Output|\n|-------|---|---|------|----|----|\n|FastPitch|fastpitch|Text-\u003eMel|46.3M|[arxiv](https://arxiv.org/abs/2006.06873)|Mel (80 bins)|\n|MixerTTS|mixer128|Text-\u003eMel|2.9M|[arxiv](https://arxiv.org/abs/2110.03584)|Mel (80 bins)|\n|MixerTTS|mixer80|Text-\u003eMel|1.5M|[arxiv](https://arxiv.org/abs/2110.03584)|Mel (80 bins)|\n|HiFi-GAN|hifigan|Vocoder|13.9M|[arxiv](https://arxiv.org/abs/2010.05646)|Wave (22.05kHz)|\n|Vocos|vocos|Vocoder|13.4M|[arxiv](https://arxiv.org/abs/2306.00814)|Wave (22.05kHz)|\n|Vocos|vocos44|Vocoder|14.0M|[arxiv](https://arxiv.org/abs/2306.00814)|Wave (44.1kHz)|\n\nThe sequence of transformations is as follows:\n\n*Text* \u0026rarr; Phonemizer \u0026rarr; *Phonemes* \u0026rarr; Tokenizer \u0026rarr; *Token Ids* \u0026rarr; **Text-\u003eMel** model \u0026rarr; *Mel spectrogram* \u0026rarr; **Vocoder** model \u0026rarr; *Wave*\n\nThe `Text-\u003eMel` models map token ids to mel frames. All models use the 80 bin configuration proposed by [HiFi-GAN](https://github.com/jik876/hifi-gan). This mel spectrogram contains frequencies up to 8kHz. The `vocoder` models map the mel spectrogram to a waveform. The vocoders with `vocoder_id` `hifigan` and `vocos` artificially extend the bandwidth to 11025Hz, and `vocos44` to 22050Hz. Samples for comparing the models can be found [here](https://nipponjo.github.io/tts-arabic-speakers/#models-cmp).\n\n**Manuscript**\n\nMore information about how the models were trained can be found in the manuscript *Arabic TTS with FastPitch: Reproducible Baselines, Adversarial Training, and Oversmoothing Analysis* ([arXiv](https://arxiv.org/abs/2512.00937) | [ResearchGate](https://www.researchgate.net/publication/398225096_Arabic_TTS_with_FastPitch_Reproducible_Baselines_Adversarial_Training_and_Oversmoothing_Analysis)).\n\n\n**TTS options**\n```python\nfrom tts_arabic import tts\n\ntext = \"اَلسَّلامُ عَلَيكُم يَا صَدِيقِي.\"\nwave = tts(\n    text, # input text\n    speaker = 1, # speaker id; choose between 0,1,2,3\n    pace = 1, # speaker pace\n    denoise = 0.005, # vocoder denoiser strength\n    volume = 0.9, # Max amplitude (between 0 and 1)\n    play = True, # play audio?\n    pitch_mul = 1, # pitch multiplier\n    pitch_add = 0, # pitch offset\n    vowelizer = None, # vowelizer model\n    model_id = 'fastpitch', # Model ID for Text-\u003eMel model\n    vocoder_id = 'hifigan', # Model ID for vocoder model\n    cuda = None, # Optional; CUDA device index\n    save_to = './test.wav', # Optionally; save audio WAV file\n    bits_per_sample = 32, # when save_to is specified (8, 16 or 32 bits)\n    )\n\n```\n\n**Vowelizer models**\n\n|Model|Model ID|Paper|Repo|Architecture|\n|-----|--------|---------|----|--|\n|CATT|catt_eo|[arxiv](https://arxiv.org/abs/2407.03236)|[github](https://github.com/abjadai/catt)|Transformer Encoder|\n|Shakkelha|shakkelha|[arxiv](https://arxiv.org/abs/1911.03531)|[github](https://github.com/AliOsm/shakkelha)|Bi-LSTM|\n|Shakkala|shakkala|-|[github](https://github.com/Barqawiz/Shakkala)|Bi-LSTM|\n\n\n**References**\n\nThe vocoder `vocos44` was converted from ([patriotyk/vocos-mel-hifigan-compat-44100khz](https://huggingface.co/patriotyk/vocos-mel-hifigan-compat-44100khz)).\n\nThe vowelizer `catt_eo` was converted from https://github.com/abjadai/catt/releases/tag/v2 *best_eo_mlm_ns_epoch_193.pt* (License: [Apache-2.0](https://github.com/abjadai/catt?tab=Apache-2.0-1-ov-file#readme))\n\n\n![DALL·E 2025-03-14 18 56 01 - A surreal digital painting of a camel in a vast desert, with a futuristic speaker embedded in its mouth, symbolizing text-to-speech technology  The ca](https://github.com/user-attachments/assets/bcd31436-1e76-4432-9072-d0695f4d87e0)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnipponjo%2Ftts_arabic","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnipponjo%2Ftts_arabic","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnipponjo%2Ftts_arabic/lists"}