FOSS AI and voice technology specialists. Creators of the HiveMind stack and core contributors to OpenVoiceOS, building privacy-first, GDPR-compliant voice solutions.https://tigregotico.pt/https://tigregotico.pt/blog/2026-08-01-exporting-and-quantizing-open-speech-models/https://tigregotico.pt/blog/2026-08-01-exporting-and-quantizing-open-speech-models/A trained speech model on a research GitHub page is not a voice assistant. We convert open ASR and TTS checkpoints to ONNX, CoreML and GGUF, quantize them, and validate the output. Then we publish the results under OpenVoiceOS so every language they cover ships in a real, offline assistant.Sat, 01 Aug 2026 00:00:00 GMTONNXCoreMLGGUFASROVOSOpenVoiceOSQuantizationOpen Sourcehttps://tigregotico.pt/blog/2026-08-02-porting-with-machines/https://tigregotico.pt/blog/2026-08-02-porting-with-machines/We rewrote several C, C++ and Java programs (espeak-ng's G2P, Cotovia, AhoTTS, HermiT) as pure Python, with an AI reading the original source and a human orchestrating. No human on our side read the originals. That raises two separate questions: can the output be owned at all, and is it a derivative of the input? We kept the upstream licences because that was cheaper than answering. We still think the question is open.Sat, 01 Aug 2026 00:00:00 GMTFOSSLicensingOpen SourcePythonG2Phttps://tigregotico.pt/blog/2026-08-03-a-family-of-pure-onnx-speech-libraries/https://tigregotico.pt/blog/2026-08-03-a-family-of-pure-onnx-speech-libraries/TigreGótico maintains a set of speech libraries — bandwidth extension, voice cloning, speaker embeddings, VAD, word stress, phonemization, TTS, and a metrics library to score them all — that share one runtime rule: onnxruntime and numpy only, no PyTorch, no GPU required.Sat, 01 Aug 2026 00:00:00 GMTONNXTTSvoice cloningVADself-hostedphoonnxhttps://tigregotico.pt/blog/2026-08-10-the-phonology-stack/https://tigregotico.pt/blog/2026-08-10-the-phonology-stack/An architecture tour of our text-to-pronunciation stack: scriptconv for notation, orthography2ipa as the cross-language grapheme-to-IPA engine, language-specific frontends built on top of it for Portuguese, Basque, Mirandese, Barranquenho and Arabic, and phonematcher for sound-based search. Shows why the layers exist, what a candidate lattice is, and real dialect output.Sat, 01 Aug 2026 00:00:00 GMTG2PIPAPhoneticsNLPTTSLinguisticsFOSShttps://tigregotico.pt/blog/2026-08-17-choosing-a-voice-cloning-engine/https://tigregotico.pt/blog/2026-08-17-choosing-a-voice-cloning-engine/voiceclonnx runs 10 voice-conversion engines behind one API, from kNN feature-swap to AR codec-LM. This is a guide to the model families behind them, the real measured trade-off between intelligibility and speaker similarity, and how to pick an engine for a specific job.Sat, 01 Aug 2026 00:00:00 GMTONNXvoice cloningvoice conversionself-hostedhttps://tigregotico.pt/blog/2026-08-24-cleaning-up-bad-audio/https://tigregotico.pt/blog/2026-08-24-cleaning-up-bad-audio/audiosronnx treats denoising and bandwidth extension as two separate jobs: the real engine registry, model sizes and licenses, the rejected-models list, and how to check whether the output actually improved.Sat, 01 Aug 2026 00:00:00 GMTONNXdenoisingbandwidth extensionspeechhttps://tigregotico.pt/blog/2026-08-31-measuring-speech-quality/https://tigregotico.pt/blog/2026-08-31-measuring-speech-quality/A working guide to speechonnxmetrics: what MOS, no-reference MOS predictors, intrusive signal metrics and ASR-based WER/CER actually measure, when each applies, real scores from real audio, and why a predicted MOS is evidence, not truth.Sat, 01 Aug 2026 00:00:00 GMTspeechonnxmetricsTTSONNXevaluationhttps://tigregotico.pt/blog/2026-09-07-scripts-and-phonetic-notations/https://tigregotico.pt/blog/2026-09-07-scripts-and-phonetic-notations/A tour of scriptconv, the zero-dependency library that detects writing systems and converts between phonetic notations. Covers IPA, ARPABET, and X-SAMPA, ISO-15924 script detection, Buckwalter transliteration for Arabic, Hangul decomposition into jamo, and kana conversion, with real, executed examples and honest limits.Sat, 01 Aug 2026 00:00:00 GMTIPAPhoneticsNLPLinguisticsFOSShttps://tigregotico.pt/blog/2026-07-13-complaint-or-goodbye/https://tigregotico.pt/blog/2026-07-13-complaint-or-goodbye/Two angry-looking support messages. One customer is about to escalate; the other is about to leave without a word. Almost no emotion model can tell them apart — because they're all missing the same axis. Introducing emotion-algebra.Mon, 13 Jul 2026 00:00:00 GMTAffective ComputingEmotionMachine LearningLILACSOpen SourceSciencehttps://tigregotico.pt/blog/2026-07-04-from-scraped-data-to-smarter-models/https://tigregotico.pt/blog/2026-07-04-from-scraped-data-to-smarter-models/Clean, typed, well-provenanced data is the raw material of every model we ship. How the catalogues our scrapers build become ASR biasing vocabularies, intent classifiers, synthetic NER corpora, G2P lexicons, TTS voices, and honest fuel for LLMs.Sat, 04 Jul 2026 00:00:00 GMTDatasetsData CollectionASRNLPTTSLLMFOSShttps://tigregotico.pt/blog/2026-07-04-turning-websites-into-voice-apps/https://tigregotico.pt/blog/2026-07-04-turning-websites-into-voice-apps/Every site that matters got wrapped in a mobile app. We propose the opposite move for the voice and CLI era: a clean API plus a voice skill per site, so the web becomes browsable by ear and by keyboard. One site at a time, it adds up to a voice browser.Sat, 04 Jul 2026 00:00:00 GMTVoiceAccessibilityOpenVoiceOSWeb AutomationCLIFOSShttps://tigregotico.pt/blog/2026-07-01-usenet-and-remailers-in-2026/https://tigregotico.pt/blog/2026-07-01-usenet-and-remailers-in-2026/Usenet is a pristine archive of pre-AI human discourse: decades of newsgroup posts, all human-written, none of it touched by language models. That makes it valuable training and evaluation data for language and speech models. We built a small Python tool to harvest it.Wed, 01 Jul 2026 00:00:00 GMTUsenetDatasetsNLPhttps://tigregotico.pt/blog/2026-06-15-two-voices-every-language-miro-and-dii/https://tigregotico.pt/blog/2026-06-15-two-voices-every-language-miro-and-dii/TigreGótico is partnering with OpenVoiceOS to give the assistant two consistent voice identities, Miro and Dii, that sound the same in every language, built with our voice-cloning technology and the phoonnx engine. Two TTS models for every language someone requests, endangered tongues included.Mon, 15 Jun 2026 00:00:00 GMTphoonnxTTSOVOSvoice cloningG2Planguage inclusionhttps://tigregotico.pt/blog/2026-06-12-disambiguating-portuguese-heterographs-for-tts/https://tigregotico.pt/blog/2026-06-12-disambiguating-portuguese-heterographs-for-tts/Many European Portuguese words are spelled the same but pronounced differently depending on meaning, and getting the vowel wrong makes a TTS voice say the wrong word. We built bifonia-pt-homographs, an open meaning-labelled dataset of 56,891 sentences over 27 words, and a tiny zero-dependency resolver that hits ≈94% where heavyweight POS taggers plateau at ≈75%.Fri, 12 Jun 2026 00:00:00 GMTDatasetsPortugueseTTSGrapheme-to-PhonemeNLPAccessibilityFOSShttps://tigregotico.pt/blog/2026-05-10-tts-that-runs-on-a-potato/https://tigregotico.pt/blog/2026-05-10-tts-that-runs-on-a-potato/phoonnx is a research framework for VITS-based text-to-speech built to run comfortably on low-end hardware. No GPU, no cloud, no API key: just a ~15.65 million-parameter ONNX voice and a CPU. Here is how small a good voice can be, and how we train them.Sun, 10 May 2026 00:00:00 GMTphoonnxTTSONNXVITSself-hostedOVOShttps://tigregotico.pt/blog/2026-04-20-music-database-scrapers/https://tigregotico.pt/blog/2026-04-20-music-database-scrapers/A tour of the family of typed Python clients we maintain for music sources — Bandcamp, SoundCloud, SomaFM, TuneIn, iHeartRadio, and the great music encyclopedias — all emitting consistent, typed media metadata behind one clean interface and riding the same compliant, low-volume HTTP transport.Mon, 20 Apr 2026 00:00:00 GMTScrapersMedia MetadataMusicPythonFOSShttps://tigregotico.pt/blog/2026-04-01-multilingual-sentence-types-dataset/https://tigregotico.pt/blog/2026-04-01-multilingual-sentence-types-dataset/We published sentence-types-multilingual — nearly 70,000 sentences across seven languages, classified by grammatical type (question, command, statement, exclamation). It is the training corpus behind the little_questions routing library.Wed, 01 Apr 2026 00:00:00 GMTDatasetsMultilingualNLPIntentClassificationFOSShttps://tigregotico.pt/blog/2026-03-15-beating-bot-walls-with-drop-in-requests-sessions/https://tigregotico.pt/blog/2026-03-15-beating-bot-walls-with-drop-in-requests-sessions/Two composable requests.Session subclasses for reading public web pages reliably without a headless browser in the hot path: TLS-compatible transport, a FlareSolverr proxy for JS challenges, a Wayback Machine fallback, and IP-diversified requests: unblock_requests and anon_requests.Sun, 15 Mar 2026 00:00:00 GMTHTTPScrapingCloudflareAnti-BotPythonOpen Sourcehttps://tigregotico.pt/blog/2026-03-01-robot-txt-sitemaps-ethical-web-scraping/https://tigregotico.pt/blog/2026-03-01-robot-txt-sitemaps-ethical-web-scraping/Before you build a scraper, scout the site. sitemapper reads robots.txt, fetches every sitemap, and optionally crawls the link graph, so your scraper starts from the site's own contract instead of brute force.Sun, 01 Mar 2026 00:00:00 GMTWeb ScrapingSitemapsEthicsRobots.txtData CollectionFOSShttps://tigregotico.pt/blog/2026-02-28-classical-nlp-for-portuguese-syllables-and-phonemes/https://tigregotico.pt/blog/2026-02-28-classical-nlp-for-portuguese-syllables-and-phonemes/A look at our rule-based, fully offline Portuguese NLP stack: silabificador for syllabification and TugaPhone for dialect-aware grapheme-to-phoneme, and how they connect to the broader orthography2ipa work for Lusophone varieties. No deep-learning black boxes: deterministic, fast, and dependency-light.Sat, 28 Feb 2026 00:00:00 GMTNLPPortuguesePhonemizationGrapheme-to-PhonemeLusophoneFOSShttps://tigregotico.pt/blog/2026-01-15-grapheme-to-ipa-for-350-languages/https://tigregotico.pt/blog/2026-01-15-grapheme-to-ipa-for-350-languages/orthography2ipa is a pure-data, linguistically grounded resource that maps spelling to IPA and models how phonemes surface as allophones across 909 language specs, 820 languages, and 20+ language families. A candidate lattice, a maximal-munch tokenizer, phonological and script distance metrics, dialect lineage, and a schema-validated spec set cited to the dialectological literature — no trained weights, fully self-hostable.Thu, 15 Jan 2026 00:00:00 GMTG2PIPAPhoneticsNLPTTSASRLinguisticsFOSShttps://tigregotico.pt/blog/2025-12-12-barranquenho/https://tigregotico.pt/blog/2025-12-12-barranquenho/g2p_barranquenho is the first open grapheme-to-phoneme converter for Barranquenho, the Ibero-Romance contact language of Barrancos, Portugal — rules derived from the municipality's newly published orthographic convention, auditable against the committed sources.Fri, 12 Dec 2025 00:00:00 GMTPhonemizationBarranquenhoMinority LanguagesNLPhttps://tigregotico.pt/blog/2025-12-09-ast/https://tigregotico.pt/blog/2025-12-09-ast/We are releasing experimental text-to-speech models for Asturian (ast) and Aragonese (an), two minority Romance languages with essentially no commercial voice coverage. We built them with a hybrid pipeline: filtered Common Voice data, zero-shot revoicing, and VITS training via phoonnx.Tue, 09 Dec 2025 00:00:00 GMTTTSAsturianAragoneseMinority LanguagesVoice Cloninghttps://tigregotico.pt/blog/2025-11-26-ovos-hivemind-industry/https://tigregotico.pt/blog/2025-11-26-ovos-hivemind-industry/The COALA and WASABI EU projects have built an entire industrial voice-assistant framework around OVOS + HiveMind, integrating them with their own tools, UI, and conversation engines.Wed, 26 Nov 2025 00:00:00 GMTOVOSHiveMindIndustrymanufacturinghttps://tigregotico.pt/blog/2025-10-14-synthetic-wakeword-datasets/https://tigregotico.pt/blog/2025-10-14-synthetic-wakeword-datasets/We published seven synthetic wakeword datasets for common voice assistant names: hey_computer, hey_mycroft, hey_siri, alexa, home_assistant, voice_assistant, wake_up. Train a detector that works everywhere.Tue, 14 Oct 2025 00:00:00 GMTDatasetsWakewordsSpeechSyntheticVoiceFOSShttps://tigregotico.pt/blog/2025-10-06-phoonnx/https://tigregotico.pt/blog/2025-10-06-phoonnx/phoonnx is now the primary text-to-speech framework for OpenVoiceOS: a full VITS/ONNX training and inference stack, starting with new Basque voices for Miro and Dii.Mon, 06 Oct 2025 00:00:00 GMTphoonnxTTSOVOSONNXPhonemizationhttps://tigregotico.pt/blog/2025-09-17-ovos_ha_dream_team/https://tigregotico.pt/blog/2025-09-17-ovos_ha_dream_team/Home Assistant handles automation. OVOS handles voice. Three integration layers make the combination work: Wyoming bridges for HA's voice pipeline, ovos-persona-server as a conversational agent, and HiveMind for surfacing OVOS devices as native HA entities.Wed, 17 Sep 2025 00:00:00 GMTOVOSHome AssistantSmart HomeVoice Automationhttps://tigregotico.pt/blog/2025-06-26-making-synthetic-voices-from-scratch/https://tigregotico.pt/blog/2025-06-26-making-synthetic-voices-from-scratch/Creating a voice for a text-to-speech system usually requires a real person to spend hours recording audio. That’s expensive, time-consuming, and in many languages or accents, the voices just don’t exist at allThu, 26 Jun 2025 00:00:00 GMTTTSSynthetic DataVoice CloningOVOShttps://tigregotico.pt/blog/2025-06-23-miro-dii-tts-training-datasets-launch/https://tigregotico.pt/blog/2025-06-23-miro-dii-tts-training-datasets-launch/We published 40 synthetic training datasets for the Miro and Dii voices across Portuguese, Dutch, German, French, Italian, Japanese, Spanish, and more. Every language gets two consistent voice identities, built with voice cloning.Mon, 23 Jun 2025 00:00:00 GMTTTSVoiceDatasetsMiro & DiiMultilingualFOSShttps://tigregotico.pt/blog/2023-10-16-no-language-left-behind/https://tigregotico.pt/blog/2023-10-16-no-language-left-behind/Eliminating language barriers in OpenVoiceOS through language detection, translation plugins, and bidirectional translation capabilities.Mon, 16 Oct 2023 00:00:00 GMTOVOSmultilinguallanguage-detectiontranslationaccessibility