{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T12:39:15.530Z","data":{"id":"brf_yGawGU_kFvMc","kind":"daily","slug":"2026-10-18","window":{"from":"2026-10-18T00:00:00.000Z","to":"2026-10-19T00:00:00.000Z"},"version":1,"quiet":false,"title":"AI brief · 18 October 2026","headline":{"text":"Whistle can now be tuned for a speaker or new language","recordIds":["wsa_6p9gfcxrrshe"]},"dek":{"text":"Whistle adds local fine-tuning, so developers can adapt its on-device speech transcription to a speaker, vocabulary, or new language. SupraTTS-1.0-Preview generates English speech locally from text on a CPU, without needing a network connection at runtime.","recordIds":["wsa_6p9gfcxrrshe","wsa_gnce4qrsnkm9"]},"tldr":[{"text":"Whistle adds local fine-tuning, so developers can adapt its on-device speech transcription to a speaker, vocabulary, or new language.","recordIds":["wsa_6p9gfcxrrshe"]},{"text":"SupraTTS-1.0-Preview generates English speech locally from text on a CPU, without needing a network connection at runtime.","recordIds":["wsa_gnce4qrsnkm9"]}],"stories":[{"blocks":[{"kind":"paragraph","items":[{"text":"Cactus Compute’s **Whistle** speech recognition model now supports local fine-tuning: developers can adapt it to particular speakers or vocabulary, or teach it a new language. The tuned model can be exported as a .cact file.","recordIds":["wsa_6p9gfcxrrshe"]},{"text":"Whistle runs on-device on a CPU and can transcribe speech in seven languages, add word timestamps, and generate speech embeddings. These features may be useful for apps and devices that need speech processing locally, including phones, wearables, and smart-home devices.","recordIds":["wsa_6p9gfcxrrshe"]}]}],"heading":"Whistle can now be tuned for a speaker or new language"},{"blocks":[{"kind":"paragraph","items":[{"text":"**SupraTTS-1.0-Preview** is an open-weight text-to-speech model that turns English text into audio and is designed to run locally with a Python runtime. It can synthesize speech on a CPU without requiring a network connection at runtime.","recordIds":["wsa_gnce4qrsnkm9"]},{"text":"That may suit developers building tools that need offline English speech generation; the record describes it as a preview release.","recordIds":["wsa_gnce4qrsnkm9"]}]}],"heading":"SupraTTS can create English speech without a network connection"}],"radar":[],"sections":[{"heading":"Models","category":"model","recordIds":["wsa_6p9gfcxrrshe","wsa_gnce4qrsnkm9"],"sentences":[]}],"items":[{"recordId":"wsa_6p9gfcxrrshe","versionId":"ver_3HZkGFhWwerw","eventType":"major_release","category":"model","title":"Whistle","maker":"Cactus Compute","summary":"Whistle is a 16.9 MB speech recognition model that runs on-device on a CPU. It transcribes speech in seven languages and can also return word timestamps and speech embeddings.","releaseNote":"Adds local fine-tuning for adapting Whistle to speakers and vocabulary or teaching it a new language, with export of the tuned model as a .cact file.","classifications":{"developmentType":"model","capability":["extraction","embeddings","fine-tuning"],"modality":["audio","text"],"access":["SDK","CLI","open weights"],"deployment":["on-device","on-premises"],"pricing":["unknown"],"targetUser":["developer","researcher"]},"citations":[{"url":"https://cactuscompute.com/blog/whistle","authority":"owner-controlled","publisher":"Cactus Compute"},{"url":"https://huggingface.co/Cactus-Compute/whistle","authority":"owner-controlled","publisher":"Hugging Face"},{"url":"https://cactuscompute.com/blog/finetuning-whistle","authority":"owner-controlled","publisher":"Cactus Compute"}],"score":0.946,"confidence":"high","correctionState":"current","publishedAt":"2026-10-02T00:00:00.000Z","effectiveAt":"2026-10-18T16:58:49.971Z","url":"https://wowsuchai.withglint.app/records/wsa_6p9gfcxrrshe","apiUrl":"https://wowsuchai.withglint.app/api/v1/records/wsa_6p9gfcxrrshe","webUrl":"https://wowsuchai.withglint.app/records/wsa_6p9gfcxrrshe"},{"recordId":"wsa_gnce4qrsnkm9","versionId":"ver__pINzb2rTLjf","eventType":"new_development","category":"model","title":"SupraTTS-1.0-Preview","maker":"SupraLabs","summary":"SupraTTS-1.0-Preview is an open-weight English text-to-speech model with 1,031,754 parameters. It generates 24 kHz mono audio and is designed to run locally with a Python runtime.","releaseNote":null,"classifications":{"developmentType":"model","capability":["audio generation"],"modality":["text","audio"],"access":["open weights","CLI"],"deployment":["on-device"],"pricing":["free","open-source"],"targetUser":["developer","researcher"]},"citations":[{"url":"https://huggingface.co/SupraLabs/SupraTTS-1.0-Preview","authority":"owner-controlled","publisher":"Hugging Face"}],"score":0.86,"confidence":"high","correctionState":"current","publishedAt":"2026-10-15T20:13:28.000Z","effectiveAt":"2026-10-18T04:54:44.934Z","url":"https://wowsuchai.withglint.app/records/wsa_gnce4qrsnkm9","apiUrl":"https://wowsuchai.withglint.app/api/v1/records/wsa_gnce4qrsnkm9","webUrl":"https://wowsuchai.withglint.app/records/wsa_gnce4qrsnkm9"}],"corrections":[],"notes":[],"writer":{"kind":"model","model":"openai/gpt-6-luna"},"publishedAt":"2026-10-19T00:35:11.244Z","updatedAt":"2026-10-19T00:35:11.244Z","url":"https://wowsuchai.withglint.app/briefs/daily/2026-10-18","apiUrl":"https://wowsuchai.withglint.app/api/v1/briefs/daily/2026-10-18"},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}