Catalogue: models and datasets
Every model and every dataset Hebrish publishes, each linking to its own page with its access state: direct download, accept the terms (approved instantly), or needs approval (a person reviews each request).
Models (12)
Access shows the main repository of each model; formats and older versions can differ, and each model page lists them. Compare models.
| Name | Type | Access | What it is |
|---|---|---|---|
| Sabich | Speech recognition | Needs approval | Parakeet fine-tune: code-switching plus tech. Hugging Face |
| Sabich-Qwen3-ASR-1.7B | Speech recognition | Needs approval | Proof of concept: Sabich's training data on Qwen3-ASR 1.7B, a one-off experiment comparing a Qwen3-ASR base with the Parakeet base. Hugging Face |
| Sabich-Canary-Qwen-2.5B | Speech recognition | Needs approval | Sabich on Canary-Qwen 2.5B: the strongest plain English of the Sabich variants, for desktop GPUs. Hugging Face |
| Sabich-Parakeet-110M | Speech recognition | Needs approval | Sabich on Parakeet 110M: for low-end phones, watches and wearables. Hugging Face |
| Sabich-asr4all | Speech recognition | Needs approval | Sabich on asr4all S and M: the smallest, streaming, for wearables and very low-power devices. Hugging Face |
| Jachnun | Speech recognition | Needs approval | Parakeet fine-tune: English with Hebrew words. Hugging Face |
| Shakshuka | Speech recognition | Needs approval | Nemotron streaming fine-tune: text appears while you speak. Hugging Face |
| Whisper Hebrish Small ACFT | Speech recognition | Accept terms | Whisper Small fine-tune for FUTO Keyboard. Hugging Face |
| Whisper Hebrish | Speech recognition | Needs approval | The first Hebrish model, on Whisper large-v3-turbo. Hugging Face |
| Hebrew-Latin Token Classifier | Small model (token classifier) | Direct download | Finds Hebrew words written in Latin letters inside English text. Hugging Face |
| Meorav | Text-to-speech | Accept terms | Text-to-speech: Irish-accented English that says the Hebrew words correctly. Hugging Face |
| Fritter | Speech recognition | Direct download | Daniel's English dictation model, no code-switching. Not a Hebrish model. Hugging Face |
Datasets (67)
Mirrored from the dataset index, built 2026-11-06. Datasets grouped by kind.
| Name | Type | Access | What it is |
|---|---|---|---|
| Acronym-Sentences | Audio dataset | Direct download | Natural sentences with acronyms read by one speaker, both spelled out (API, SSH, HTTPS) and said as words (NASA, JSON, YAML), each in its standard written form Hugging Face |
| AI-Interactions-Sentence-Bank | Sentence dataset | Direct download | Sentences dictated to AI agents (text only). Hugging Face |
| AI-Interactions-Sentences | Audio dataset | Direct download | Instructions dictated to AI coding agents (Claude Code, Codex and others) and the git and GitHub workflow around them, read aloud by one speaker: subagents, worktrees, compaction, plan mode, hooks, MCP servers, pull requests, rebases Hugging Face |
| AI-Tech-Vocabulary-Sentence-Bank | Sentence dataset | Accept terms | AI and tech vocabulary sentence bank (text only). Hugging Face |
| Annotated-Voice-Notes | Related (outside the Hebrish speech models) | Direct download | Real voice notes with annotations. Hugging Face |
| ASR-Cleanup-Before-After | Related (outside the Hebrish speech models) | Direct download | ASR cleanup before and after pairs. Separate project. Hugging Face |
| ASR-Error-List | Error list | Accept terms | Words the models mishear in Daniel's own dictation, with the wrong forms; drives the capture app's priority words. Sabich v2.1 repeated its takes 3 times. Hugging Face |
| ASR-Error-Prediction | Error list | Direct download | Predicted mistranscription candidates for planning new sentences. Hugging Face |
| ASR-Paragraph-Breaks | Related (outside the Hebrish speech models) | Direct download | Hand-annotated paragraph breaks for flat speech-to-text output. Separate project. Hugging Face |
| Background-Noise | Reference data | Accept terms | Background noise recordings. Hugging Face |
| Brand-Name-Sentences | Audio dataset | Accept terms | Natural sentences with company, product, AI model and open-source project names (Anthropic, OpenRouter, Amazon Bedrock, Qwen, Kubernetes, Nginx) read by one speaker, written in each name's official form Hugging Face |
| Community-Voice-Contributions | Audio dataset | Direct download | Recordings contributed by anyone, for the speech models. Hugging Face |
| Currency-Sentences | Audio dataset | Accept terms | One speaker saying amounts of money the natural way ("50 shekels", "a dollar fifty", "6 pounds 50") Hugging Face |
| Disfluent-Read-Sentences | Audio dataset | Accept terms | Experimental Hugging Face |
| English-Hebrew-Codeswitching-ASR-Benchmark | Benchmark | Accept terms | Held-out test sets (v1, v2) the models are scored on; never trained on. Hugging Face |
| English-Hebrew-Codeswitching-Sentences | Audio dataset | Accept terms | English sentences with everyday Hebrew words and Israeli names written in Latin letters, read aloud by one English speaker living in Israel Hugging Face |
| English-Hebrew-Mixed-Sentences | Audio dataset | Needs approval | The first code-switching set: sentences with recordings (637 WAV files, checked 2026-11-05). Its sentences are also a source of the prompt bank. Hugging Face |
| English-Hebrew-Mixed-Sentences | Other | Direct download | Card only, no data files (checked 2026-11-05). See Hebrish/English-Hebrew-Mixed-Sentences. Hugging Face |
| Example-Audio-Sentences | Reference data | Accept terms | Format reference for audio datasets: the project's standard columns, with a schema file and three real takes with the place block blanked. Hugging Face |
| Example-Sentence-Bank | Reference data | Accept terms | Format reference for sentence datasets: the project's standard columns, with a schema file and a handful of invented neutral sentences. Hugging Face |
| Expanded-Acronym-Sentences | Audio dataset | Accept terms | Chat and business shorthand said as letters by one speaker ("by E O D", "F Y I", "I M O") with the transcript writing it out in full ("by end of day", "For your information", "In my opinion") Hugging Face |
| General-English-Sentences | Audio dataset | Direct download | Plain, everyday English sentences (emails, messages, general prose and homophone pairs) read aloud by one speaker, with no Hebrew or Yiddish words, Israeli names or tech vocabulary Hugging Face |
| Hard-ASR-Sentences | Audio dataset | Accept terms | English sentences speech recognition finds hard, read by one speaker: Harvard sentences (IEEE 1969, public domain), tongue twisters, homographs, garden-path sentences, rare words, dense numbers and units, unusual proper nouns Hugging Face |
| hebrew-latin-code-switching | Sentence dataset | Accept terms | Sentences with romanized Hebrew words labelled, for the token classifier. Hugging Face |
| Hebrew-Term-Pronunciations-IPA | Reference data | Accept terms | How Hebrew words are pronounced inside English speech, in IPA, measured from real recordings. Hugging Face |
| Hebrish-Ideas | Other | Direct download | Ideas recorded in the capture app. Hugging Face |
| Hebrish-TTS-Voice-Recordings | Audio dataset | Accept terms | Hugging Face |
| Hebrish-Validation | Benchmark | Direct download | Validation benchmark for the English-Hebrew code-switching speech models. Hugging Face |
| Homophone-Context-Sentence-Bank | Sentence dataset | Accept terms | Homophone-in-context sentence bank (text only). Hugging Face |
| Homophone-Context-Sentences | Audio dataset | Direct download | English sentences whose context settles which spelling of a homophone is meant (their/there, principal/principle, and 58 more sets), read aloud by one speaker Hugging Face |
| Irish-English-Sentences | Audio dataset | Accept terms | Everyday sentences in Irish English read by one speaker born in Ireland: vocabulary (craic, grand, giving out, press, messages), grammar (I'm after doing it, so it is, ye), Irish-language words used in English (Taoiseach, Dáil, Gardaí, slái Hugging Face |
| Israeli-Name-Sentences | Audio dataset | Accept terms | English messages and emails that mention an Israeli or Jewish person's name, a first name (Naftali, Yotam, Tzlil, Bat-El) or a surname (Buskila, Ifergan, Weinberg), written in its common Israeli-English spelling, read aloud by one speaker Hugging Face |
| Israeli-Street-Names | Reference data | Accept terms | Israel's national street list, source for the street-name prompts. Hugging Face |
| Jewish-Religious-Terminology-Sentences | Audio dataset | Needs approval | English sentences with Jewish religious vocabulary written in Latin letters: Shabbat and the festivals, prayer and the synagogue, Torah portions, Talmud and halacha, kashrut, and life-cycle events, read aloud by one speaker Hugging Face |
| Judaica-Sentences | Audio dataset | Direct download | Real sentences from traditional Jewish works in English, read by one speaker, Hebrew terms in romanized form: Tanach (Metsudah), Mishnah (Mishnah Yomit, Joshua Kulp; Shraga Silverstein; Sefaria Community Translation; Eighteen Treatises from Hugging Face |
| Linux-Admin-Sentence-Bank | Sentence dataset | Accept terms | Linux administration sentence bank (text only). Hugging Face |
| Linux-Admin-Sentences | Audio dataset | Direct download | Dictation-style English sentences about Linux desktop and server administration (KDE Plasma, GNOME, systemd, storage, networking, homelab), read aloud by one speaker Hugging Face |
| Model-Checks | Audio dataset | Accept terms | Takes checked by a speech recognition model while they were recorded, with the model's transcript, the speaker's verdict on it and, once judged, a second, machine verdict Hugging Face |
| Noisy-Condition-Sentences | Audio dataset | Accept terms | Every take from the main dataset recorded in noise, whatever its category: outdoors, in a noisy room, with traffic, other people talking, background noise or a crying baby Hugging Face |
| October-7-Memorial-Sentence-Bank | Sentence dataset | Accept terms | In memory of those who were murdered and who fell on October 7, 2023: three sentences about each person's life, and places and terms connected with that day. Hugging Face |
| Paragraph-Dictation-Passages | Audio dataset | Direct download | Experimental Hugging Face |
| Parakeet-V2-Blend-All | Training mix | Needs approval | Export of the `all` training mix, for reproducing Sabich runs. Hugging Face |
| Parakeet-V2-Blend-No-Hebrew | Training mix | Needs approval | The exact data of training run r7-general, Daniel's English model without code-switching (not a Hebrish model). Hugging Face |
| Personal-Vocabulary-Sentence-Bank | Sentence dataset | Accept terms | Personal vocabulary sentence bank (text only). Hugging Face |
| Personal-Vocabulary-Sentences | Audio dataset | Direct download | English sentences built around names a speech model gets wrong for one speaker: his own name, his company and his project names (Daniel Rosehill, DSR Holdings, Carrot Cake AI, the podcast My Weird Prompts and its hosts Corn and Herman Poppl Hugging Face |
| Pronunciation-Variants | Audio dataset | Accept terms | Takes in which a word was deliberately said one particular way, from the main dataset, stratified by `variant_category` Hugging Face |
| Proofed-ASR-Cleanup-Training-Data | Related (outside the Hebrish speech models) | Direct download | Audio paired with clean, proofed text, for ASR cleanup training. Separate project. Hugging Face |
| Read-Aloud-Voice-Sentences | Audio dataset | Accept terms | Every take from the capture app; all speech models train on a pinned revision of it, selected by blend. Gated. Hugging Face |
| Repeated-Word-Sentences | Audio dataset | Accept terms | Experimental Hugging Face |
| Room-Tone-Annotations | Annotations | Direct download | Room-tone recordings and labels for noise conditions. Hugging Face |
| Self-Correction-Sentences | Audio dataset | Direct download | One speaker correcting himself mid-sentence ("send it Tuesday, no, Wednesday") Hugging Face |
| Singing-Sentences | Audio dataset | Direct download | Experimental Hugging Face |
| Smart-Home-Sentence-Bank | Sentence dataset | Direct download | Smart-home sentence bank (text only). Hugging Face |
| Smart-Home-Sentences | Audio dataset | Direct download | Dictation-style English sentences about home automation (Home Assistant, Zigbee, Z-Wave, Matter, Thread, ESPHome, MQTT and device brands), read aloud by one speaker Hugging Face |
| Speakers | Reference data | Direct download | Accreditation: one row per contributor whose voice is in the project. Hugging Face |
| Spoken-Aside-Sentences | Audio dataset | Direct download | One speaker dictating a sentence and, partway through, saying something to someone in the room ("come in, Hannah") Hugging Face |
| Synthetic-Codeswitching-Voices-Pilot | Audio dataset | Direct download | 100 synthetic clips (2026-10-28): English sentences with one Hebrew term. Hugging Face |
| Take-Labels | Annotations | Direct download | Hand labels for takes in the master dataset. Hugging Face |
| Tech-Sentences | Audio dataset | Direct download | Dictation-style English sentences dense in software, DevOps, cloud, self-hosting and AI vocabulary (product names, CLI tools, acronyms, model names), read aloud by one speaker Hugging Face |
| Tech-Sentences-For-ASR-Training | Audio dataset | Direct download | Early tech sentences with audio; a prompt source for the bank. Hugging Face |
| term-context-sentences | Sentence dataset | Direct download | Training data for the term-context sentence generator (Granite 4.2 3B). Hugging Face |
| Voice-Command-Sentences | Audio dataset | Direct download | One speaker reading desktop voice commands for a KDE Plasma voice-command agent: launching, switching to and closing apps (Konsole, Kate, Dolphin, Kdenlive, Audacity), browser and terminal tabs, windows and virtual desktops ("Move this to d Hugging Face |
| Voice-Note-Audio | Related (outside the Hebrish speech models) | Direct download | Real-world voice recordings with transcripts and annotations. Hugging Face |
| Voice-Notes | Related (outside the Hebrish speech models) | Direct download | Real voice notes. Hugging Face |
| Wake-Words | Related (outside the Hebrish speech models) | Direct download | Jewish and Israeli first-name wake words, for local wake word models. Hugging Face |
| Weird-Voice-Sentences | Audio dataset | Direct download | Experimental Hugging Face |
| Yiddish-Sentences | Audio dataset | Accept terms | English sentences with Yiddish words written in Latin letters (shul, daven, cholent, mensch, freier, tachles), the way English speakers in Israel use them, read aloud by one speaker Hugging Face |