PyPI - babelscribe - Versions diffs - 0.3.1__tar.gz → 0.3.2__tar.gz - Mend

babelscribe 0.3.1tar.gz → 0.3.2tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (23) hide show

{babelscribe-0.3.1 → babelscribe-0.3.2}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: babelscribe
-Version: 0.3.1
+Version: 0.3.2
 Summary: Transcribe any audio/video in 99 languages on any GPU (AMD, NVIDIA, Intel via Vulkan; NVIDIA via CUDA; Apple via Metal) or CPU — whisper.cpp with batteries included.
 License: MIT
 Project-URL: Homepage, https://github.com/phonology024/babelscribe
@@ -251,6 +251,25 @@ them to each `v*` release. Point `BABELSCRIBE_RELEASES` at another URL to self-h
 ภาษาไทยแนะนำ `babelscribe ไฟล์.mp4 -l th --accurate` — ข้อความจาก Pathumma Whisper (NECTEC) ที่ผิดน้อยที่สุดใน FLEURS (CER 8.9% เทียบ turbo 15.9%) + เวลาจาก large-v3-turbo
 หรือเลือก Thonburian Whisper เอง: `--text-model thai-thonburian`
+## Privacy Policy
+babelscribe runs entirely on your own computer. Last updated 2026-10-06.
+- **Data collection:** none. babelscribe has no telemetry, analytics, accounts or crash reporting, and never uploads your
+  audio, video, transcripts or file names anywhere.
+- **What it processes and where:** the media file you choose is converted and transcribed locally; subtitles and text are
+  written to your disk (next to the file, or the folder you choose). Through MCP, the transcript text is returned to the AI
+  app that called the tool — what that app does with it is governed by that app's own privacy policy.
+- **Network access (downloads only):** on first use it downloads the `whisper-cli` program from this project's GitHub
+  releases, speech models from Hugging Face (`huggingface.co/ggerganov/whisper.cpp`, and for `--accurate` Thai/Hindi the
+  fine-tune's own repository), and Python packages from PyPI when installed with pip/uv. These are plain downloads;
+  no personal data is sent. GitHub, Hugging Face and PyPI see a normal download request (IP address, user agent) under
+  their own privacy policies.
+- **Storage and retention:** downloaded programs and models are cached in `~/.babelscribe` (or `BABELSCRIBE_HOME` /
+  `BABELSCRIBE_MODELS`) until you delete that folder. Outputs stay wherever they were written until you delete them.
+  babelscribe keeps no other data.
+- **Third-party sharing:** none.
+- **Contact:** open an issue at https://github.com/phonology024/babelscribe/issues
 ## Credits & licence
 MIT. Built on [whisper.cpp](https://github.com/ggml-org/whisper.cpp) (MIT) and OpenAI Whisper models (MIT).
 Fine-tunes belong to their authors: [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) by NECTEC,

{babelscribe-0.3.1 → babelscribe-0.3.2}/README.md RENAMED Viewed

@@ -1,215 +1,234 @@
-# babelscribe — free, offline speech-to-text and subtitles on any GPU (AMD, NVIDIA, Intel, Apple)
-<!-- mcp-name: io.github.phonology024/babelscribe -->
-[![PyPI](https://img.shields.io/pypi/v/babelscribe)](https://pypi.org/project/babelscribe/)
-[![License: MIT](https://img.shields.io/badge/license-MIT-green)](LICENSE)
-[![MCP server](https://img.shields.io/badge/MCP-server-blue)](#use-it-from-an-ai-app-mcp--no-terminal-needed)
-**babelscribe turns any audio or video file into subtitles (SRT, VTT) and text, in any of Whisper's 99 languages, on your
-own graphics card — AMD Radeon, NVIDIA, Intel Arc or Apple Silicon — or the CPU. No CUDA needed, no cloud, free (MIT).**
-It runs OpenAI Whisper through [whisper.cpp](https://github.com/ggml-org/whisper.cpp) with Vulkan, CUDA or Metal, and works
-from the command line, from Python, or from AI apps (Claude, Codex, Antigravity, Gemini CLI, Cursor) as an MCP server.
-```bash
-pip install babelscribe          # Python 3.9+
-babelscribe interview.mp4                 # auto-detects language, picks your best GPU -> interview.srt + interview.json
-```
-babelscribe is a thin, friendly layer on top of [whisper.cpp](https://github.com/ggml-org/whisper.cpp). It exists because
-getting fast Whisper on a non-NVIDIA card (e.g. an AMD Radeon on Windows) still means building whisper.cpp with the
-Vulkan SDK yourself. babelscribe downloads a prebuilt `whisper-cli` for your system and handles the rest.
-| Your hardware | Backend used | Prebuilt asset |
-|---|---|---|
-| AMD / NVIDIA / Intel GPU on Windows | Vulkan | `windows-x64-vulkan` |
-| NVIDIA on Windows | Vulkan (same build) | `windows-x64-vulkan` |
-| NVIDIA on Linux (alternative) | CUDA | `linux-x64-cuda` (`--flavor cuda`) |
-| AMD / NVIDIA / Intel GPU on Linux | Vulkan | `linux-x64-vulkan` |
-| Apple Silicon | Metal | `macos-arm64-metal` |
-| No usable GPU | CPU | `*-cpu` (`--flavor cpu`) |
-A 19-minute English TED talk transcribes in 48 seconds on an AMD Radeon RX 9070 XT with 1.5% word error — see the benchmark below.
-## Use it from an AI app (MCP) — no terminal needed
-babelscribe is also an [MCP](https://modelcontextprotocol.io) server, so any AI app that supports local MCP servers can
-transcribe files on **your** computer with **your** GPU. Ask in plain words — *"make Thai subtitles for the newest video in
-my Downloads"* — and the app finds the file, transcribes it, writes `.srt` / `.txt` next to it, and can then proofread names,
-translate or summarise the text for you.
-Tools: `transcribe` (file → subtitles + text), `find_media` (newest audio/video in Downloads, Videos, Desktop, …),
-`list_devices`, `list_languages_and_models`.
-**Claude Desktop (one click):** download `babelscribe.mcpb` from the [latest release](https://github.com/phonology024/babelscribe/releases/latest)
-and double-click it (or *Settings → Extensions → Install extension*).
-**Everything else** runs the same command — [uv](https://docs.astral.sh/uv/) fetches babelscribe for you:
-`uvx babelscribe mcp`
-| App | How to add it |
-|---|---|
-| Claude Code | `claude mcp add babelscribe -- uvx babelscribe mcp` |
-| OpenAI Codex CLI | `codex mcp add babelscribe -- uvx babelscribe mcp`, then in `~/.codex/config.toml` under `[mcp_servers.babelscribe]` set `tool_timeout_sec = 3600` and `startup_timeout_sec = 120` (defaults are 60 s / 10 s — too short for a long video or the first model download) |
-| Google Antigravity | agent panel → ⋯ → *MCP Servers* → *Manage MCP Servers* → *View raw config*, add the JSON below |
-| Gemini CLI | add the JSON below to `~/.gemini/settings.json` |
-| Cursor | add the JSON below to `~/.cursor/mcp.json` |
-| VS Code (Copilot) | `.vscode/mcp.json`: same entry under `"servers"` with `"type": "stdio"` |
-```json
-{
-  "mcpServers": {
-    "babelscribe": { "command": "uvx", "args": ["babelscribe", "mcp"], "timeout": 3600000 }
-  }
-}
-```
-(`timeout` is in milliseconds and only some apps read it.) Already installed with pip? Use `"command": "babelscribe", "args": ["mcp"]`.
-Web-only chat apps (e.g. grok.com, chatgpt.com) can only reach servers on the internet, not your computer, so they can't use your GPU or files.
-## Benchmark: real talks, human captions as the answer key
-Four TED / TEDx talks, scored against the **human-made captions in the spoken language** (`bench/bench.py`, reproducible).
-Error = word error rate (WER) for space-separated languages, character error rate (CER) for Japanese and Thai.
-GPU: AMD Radeon RX 9070 XT via Vulkan. *default* = `large-v3-turbo`; *accurate* = `--accurate` (see the FLEURS section).
-| Language | Talk | Length | default: time / error | `--accurate`: time / error |
-|---|---|---|---|---|
-| English | [Matt Walker — Sleep Is Your Superpower (TED)](https://youtu.be/5MuIMqhT8DM) | 19.3 min | 48 s (24x) · WER **1.5%** | 107 s (11x) · WER **1.5%** |
-| Japanese | [Kazunari Taguchi (TEDxHimi)](https://youtu.be/cjtmDEG-B7U) | 16.2 min | 47 s (20x) · CER 4.7% | 115 s (8x) · CER **4.2%** |
-| Spanish | [Adrià Solà Pastor — Cómo hablar (TEDxESIC University)](https://youtu.be/XUqrvbsTfck) | 19.9 min | 56 s (21x) · WER 13.8% | 134 s (9x) · WER 13.5% |
-| Thai | [นิติ ชัยชิตาทร — โปรดเรียกฉันด้วยนามอันแท้จริง (TEDxBangkok)](https://youtu.be/48A9SU6_bQ8) | 14.1 min | 74 s (12x) · CER 22.7% | 565 s (2x) · CER **16.4%** |
-How to read it: TED captions are edited for reading (fillers dropped, light rewording), so these numbers are an upper bound —
-most of the Spanish "errors" are the speaker's actual words versus the tidied caption. Thai is genuinely harder: fast,
-casual speech with slang; `--accurate` (Pathumma Whisper text + turbo timing) cuts its error by more than a quarter.
-Talks are used only to measure accuracy; their transcripts are not redistributed (TED content is CC BY-NC-ND).
-## Benchmark: 16 languages on FLEURS
-[Google FLEURS](https://huggingface.co/datasets/google/fleurs) test set, 50 utterances per language, human-verified
-verbatim transcripts (`bench/fleurs.py`, reproducible). Same normaliser for every language (Whisper's rule: lower-case,
-drop punctuation and non-spacing marks, numbers spelled out). WER for space-separated languages, CER for ja / zh / ko / th.
-| Language | default (turbo) | `--accurate` | model `--accurate` uses | `--accurate`, edge-trimmed* |
-|---|---|---|---|---|
-| English | 6.1% | 5.7% | large-v3, beam 5 | 5.6% |
-| Spanish | 4.1% | 4.2% | large-v3, beam 5 | **2.6%** |
-| French | 7.4% | 7.3% | large-v3, beam 5 | 7.2% |
-| German | 4.3% | 4.0% | large-v3, beam 5 | **3.3%** |
-| Portuguese | 8.5% | 7.7% | large-v3, beam 5 | 5.3% |
-| Italian | 6.7% | 6.5% | large-v3, beam 5 | **3.8%** |
-| Russian | 6.8% | 5.8% | large-v3, beam 5 | 5.6% |
-| Arabic | 11.0% | 10.6% | large-v3, beam 5 | 9.7% |
-| Hindi | 28.4% | **12.6%** | [vasista22/whisper-hindi-large-v2](https://huggingface.co/vasista22/whisper-hindi-large-v2) + turbo timing | 11.1% |
-| Indonesian | 9.8% | 7.8% | large-v3, beam 5 | 5.6% |
-| Vietnamese | 10.8% | 9.1% | large-v3, beam 5 | 9.1% |
-| Turkish | 6.2% | 6.7% | large-v3, beam 5 | 5.8% |
-| Japanese (CER) | 6.4% | 5.7% | large-v3, beam 5 | **4.6%** |
-| Chinese (CER) | 6.3% | 5.3% | large-v3, beam 5 | **4.8%** |
-| Korean (CER) | 4.1% | 3.9% | large-v3, beam 5 | **3.1%** |
-| Thai (CER) | 15.9% | **8.9%** | [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) (NECTEC) + turbo timing | 8.8% |
-\* Some FLEURS clips contain more speech than their reference transcript, so a correct model is charged for words the
-reference leaves out. *Edge-trimmed* ignores extra words before the first / after the last reference word; both numbers
-are stored by `bench/fleurs.py`. 50 utterances per language means differences under ~0.5 points are noise.
-Also measured and **not** used, because large-v3 was as good or better: large-v2 (all languages),
-PhoWhisper-large (vi 19.0%), whisper-large-v3 dialectal / code-switching Arabic fine-tunes (12.2% / 15.2%),
-Typhoon Whisper (th 11.9%), Thonburian Whisper (th 9.1% — kept as an option), Vaani Hindi (16.5%).
-## Why babelscribe (vs. what already exists)
-| | GPU on AMD / Intel | Windows, no build step | Video in, subtitles out | Long files don't loop | Better text for your language |
-|---|---|---|---|---|---|
-| **babelscribe** | ✅ Vulkan | ✅ prebuilt `whisper-cli` downloaded for you | ✅ ffmpeg bundled | ✅ `--max-context 0` by default | ✅ hybrid fine-tune text + turbo timing |
-| whisper.cpp (raw) | ✅ Vulkan — if you compile it | ❌ official releases ship no Windows Vulkan build | ❌ WAV 16 kHz only | ⚠️ you must know the flag | ❌ |
-| faster-whisper / WhisperX | ❌ GPU = NVIDIA CUDA only | ✅ pip | ✅ | ⚠️ | ⚠️ manual |
-| Cloud APIs | n/a (cloud) | ✅ | ✅ | ✅ | ❌ — and your audio leaves your machine, paid per minute |
-babelscribe does **not** replace those projects — it stands on whisper.cpp and simply removes the hard parts:
-compiling for your GPU, converting media, picking the right device, avoiding the long-file repeat bug, and combining
-a language-specific fine-tune with accurate timestamps.
-## FAQ
-**How do I run Whisper on an AMD GPU on Windows?**
-`pip install babelscribe`, then `babelscribe video.mp4`. It downloads a Vulkan build of whisper.cpp that runs on AMD Radeon
-(and Intel Arc / NVIDIA) cards on Windows and Linux — no ROCm, no CUDA, no compiling.
-**How do I make subtitles (SRT) from a video for free, offline?**
-`babelscribe video.mp4 -f srt` writes `video.srt` next to the video. Nothing is uploaded; it runs on your own computer.
-**What is the most accurate free transcription for Thai?**
-`babelscribe video.mp4 -l th --accurate` — Pathumma Whisper (NECTEC) for the text plus Whisper turbo for timing:
-CER 8.9% on Google FLEURS vs 15.9% for plain Whisper turbo. Thonburian Whisper is available too (`--text-model thai-thonburian`).
-**Can Claude / ChatGPT Codex / Gemini transcribe a video on my computer?**
-Yes — add babelscribe as an MCP server (see *Use it from an AI app*). Claude Desktop installs it with one click from
-`babelscribe.mcpb`. The AI app can then find a file, transcribe it on your GPU, and proofread, translate or summarise the text.
-**How fast is it?**
-About 20x real time with Whisper large-v3-turbo on an AMD Radeon RX 9070 XT: a 19-minute talk in 48 seconds.
-**Which languages are supported?**
-All 99 Whisper languages, with automatic language detection. FLEURS error rates for 16 of them are in the benchmark table.
-**Is it better than faster-whisper or WhisperX?**
-Those are excellent on NVIDIA GPUs; on AMD / Intel GPUs they fall back to the CPU. babelscribe's niche is any GPU, zero setup,
-and better Thai / Hindi through community fine-tunes. If you have an NVIDIA card and like Python, faster-whisper is a fine choice.
-## Languages
-All 99 languages Whisper was trained on, auto-detected or forced with `-l`:
-af am ar as az ba be bg bn bo br bs ca cs cy da de el en es et eu fa fi fo fr gl gu ha haw he hi hr ht hu hy id is it ja jw ka kk km kn ko la lb ln lo lt lv mg mi mk ml mn mr ms mt my ne nl nn no oc pa pl ps pt ro ru sa sd si sk sl sn so sq sr su sv sw ta te tg th tk tl tr tt uk ur uz vi yi yo yue zh
-Accuracy follows Whisper's own training data: excellent for high-resource languages (English, Spanish, Japanese, …),
-weaker for low-resource ones. That is what hybrid mode is for — a community fine-tune for one language can be plugged in
-with one line in `babelscribe/models.py` (Thai and Hindi so far). PRs adding fine-tunes for other
-languages are the most valuable contribution.
-## Limitations (honest)
-- Tested end to end so far on an AMD Radeon RX 9070 XT (Windows, Vulkan). CUDA, Linux and macOS builds are produced by CI; reports from those machines are welcome.
-- Hybrid mode runs two models, so it is slower (≈1 min per minute of audio with a large fine-tune on that GPU).
-- Proper nouns can still be misspelled — check names before publishing subtitles.
-## Features
-- **Any input** — mp4, mkv, mov, mp3, wav, m4a… (ffmpeg is bundled through `imageio-ffmpeg`).
-- **Any language** — `-l auto` detects it; or pass `-l th`, `-l ja`, `-l es`…
-- **Picks the right GPU** — prefers a discrete card over an integrated one; `babelscribe devices` lists them, `--device N` overrides.
-- **Long files that don't loop** — runs whisper with `--max-context 0`, which stops the classic "same sentence repeated forever" hallucination on long recordings.
-- **Hybrid mode for better spelling in your language** — community fine-tunes (e.g. Thai *Thonburian Whisper*) spell far better but often lose timestamps. `--text-model` takes the text from the fine-tune and the timing from `large-v3-turbo`, aligned character by character (works for languages without spaces), cut only at word boundaries, with the timing model filling any words the fine-tune skipped.
-- **`--accurate`** — slower, fewest errors: large-v3 with beam search, or for Thai and Hindi the best community fine-tune (hybrid). Chosen per language from the FLEURS benchmark above.
-- **Outputs** — `srt`, `vtt`, `txt`, `json` (segments with token timings).
-## Usage
-```bash
-babelscribe talk.mp4 -f srt,vtt,txt,json          # all formats
-babelscribe podcast.mp3 -l en -m large-v3          # pick language and model
-babelscribe talk.mp4 -l hi --accurate              # slower, fewest errors (best model per language)
-babelscribe vo.wav -l th --text-model thai-thonburian   # hybrid: pick a Thai fine-tune yourself
-babelscribe devices                                # GPUs whisper.cpp can see
-babelscribe models                                 # models and fine-tunes
-babelscribe talk.mp4 --bin /path/to/whisper-cli    # use your own whisper.cpp build
-```
-From Python:
-```python
-from babelscribe.api import transcribe_file
-r = transcribe_file("talk.mp4", lang="auto", formats=["srt", "txt"])   # or accurate=True
-print(r["lang"], r["device"], r["files"]); print(r["text"][:200])
-```
-Models download on first use to `~/.babelscribe/models` (`BABELSCRIBE_MODELS` to change). Hybrid fine-tunes are converted on your
-machine from their original Hugging Face repo — install `pip install "babelscribe[finetune]"` once; converted weights are never
-redistributed, so each fine-tune keeps its own licence. Thai word boundaries: `pip install "babelscribe[thai]"`.
-## Add a language fine-tune
-Add one entry to `FINETUNES` in `babelscribe/models.py` (Hugging Face repo, language code, whether it needs beam search) and open a PR.
-## Prebuilt binaries
-`.github/workflows/build-binaries.yml` builds `whisper-cli` + `whisper-quantize` for every row of the table above and attaches
-them to each `v*` release. Point `BABELSCRIBE_RELEASES` at another URL to self-host.
-## ภาษาไทย
-ใช้ผ่าน Claude Desktop ได้โดยไม่ต้องพิมพ์คำสั่ง: โหลด `babelscribe.mcpb` จากหน้า Releases แล้วดับเบิลคลิก จากนั้นพิมพ์ในแชทว่า *"ทำซับไทยให้คลิปล่าสุดในโฟลเดอร์ดาวน์โหลด"*
-ถอดเสียงจากไฟล์เสียงหรือวิดีโอได้ทุกภาษา บนการ์ดจอทุกยี่ห้อ (AMD / NVIDIA / Intel ผ่าน Vulkan, Apple ผ่าน Metal) หรือ CPU
-ภาษาไทยแนะนำ `babelscribe ไฟล์.mp4 -l th --accurate` — ข้อความจาก Pathumma Whisper (NECTEC) ที่ผิดน้อยที่สุดใน FLEURS (CER 8.9% เทียบ turbo 15.9%) + เวลาจาก large-v3-turbo
-หรือเลือก Thonburian Whisper เอง: `--text-model thai-thonburian`
-## Credits & licence
-MIT. Built on [whisper.cpp](https://github.com/ggml-org/whisper.cpp) (MIT) and OpenAI Whisper models (MIT).
-Fine-tunes belong to their authors: [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) by NECTEC,
-[Thonburian Whisper](https://huggingface.co/biodatlab/whisper-th-large-v3-combined) by biodatlab,
-[whisper-hindi-large-v2](https://huggingface.co/vasista22/whisper-hindi-large-v2) by vasista22 (Speech Lab, IIT Madras).
+# babelscribe — free, offline speech-to-text and subtitles on any GPU (AMD, NVIDIA, Intel, Apple)
+<!-- mcp-name: io.github.phonology024/babelscribe -->
+[![PyPI](https://img.shields.io/pypi/v/babelscribe)](https://pypi.org/project/babelscribe/)
+[![License: MIT](https://img.shields.io/badge/license-MIT-green)](LICENSE)
+[![MCP server](https://img.shields.io/badge/MCP-server-blue)](#use-it-from-an-ai-app-mcp--no-terminal-needed)
+**babelscribe turns any audio or video file into subtitles (SRT, VTT) and text, in any of Whisper's 99 languages, on your
+own graphics card — AMD Radeon, NVIDIA, Intel Arc or Apple Silicon — or the CPU. No CUDA needed, no cloud, free (MIT).**
+It runs OpenAI Whisper through [whisper.cpp](https://github.com/ggml-org/whisper.cpp) with Vulkan, CUDA or Metal, and works
+from the command line, from Python, or from AI apps (Claude, Codex, Antigravity, Gemini CLI, Cursor) as an MCP server.
+```bash
+pip install babelscribe          # Python 3.9+
+babelscribe interview.mp4                 # auto-detects language, picks your best GPU -> interview.srt + interview.json
+```
+babelscribe is a thin, friendly layer on top of [whisper.cpp](https://github.com/ggml-org/whisper.cpp). It exists because
+getting fast Whisper on a non-NVIDIA card (e.g. an AMD Radeon on Windows) still means building whisper.cpp with the
+Vulkan SDK yourself. babelscribe downloads a prebuilt `whisper-cli` for your system and handles the rest.
+| Your hardware | Backend used | Prebuilt asset |
+|---|---|---|
+| AMD / NVIDIA / Intel GPU on Windows | Vulkan | `windows-x64-vulkan` |
+| NVIDIA on Windows | Vulkan (same build) | `windows-x64-vulkan` |
+| NVIDIA on Linux (alternative) | CUDA | `linux-x64-cuda` (`--flavor cuda`) |
+| AMD / NVIDIA / Intel GPU on Linux | Vulkan | `linux-x64-vulkan` |
+| Apple Silicon | Metal | `macos-arm64-metal` |
+| No usable GPU | CPU | `*-cpu` (`--flavor cpu`) |
+A 19-minute English TED talk transcribes in 48 seconds on an AMD Radeon RX 9070 XT with 1.5% word error — see the benchmark below.
+## Use it from an AI app (MCP) — no terminal needed
+babelscribe is also an [MCP](https://modelcontextprotocol.io) server, so any AI app that supports local MCP servers can
+transcribe files on **your** computer with **your** GPU. Ask in plain words — *"make Thai subtitles for the newest video in
+my Downloads"* — and the app finds the file, transcribes it, writes `.srt` / `.txt` next to it, and can then proofread names,
+translate or summarise the text for you.
+Tools: `transcribe` (file → subtitles + text), `find_media` (newest audio/video in Downloads, Videos, Desktop, …),
+`list_devices`, `list_languages_and_models`.
+**Claude Desktop (one click):** download `babelscribe.mcpb` from the [latest release](https://github.com/phonology024/babelscribe/releases/latest)
+and double-click it (or *Settings → Extensions → Install extension*).
+**Everything else** runs the same command — [uv](https://docs.astral.sh/uv/) fetches babelscribe for you:
+`uvx babelscribe mcp`
+| App | How to add it |
+|---|---|
+| Claude Code | `claude mcp add babelscribe -- uvx babelscribe mcp` |
+| OpenAI Codex CLI | `codex mcp add babelscribe -- uvx babelscribe mcp`, then in `~/.codex/config.toml` under `[mcp_servers.babelscribe]` set `tool_timeout_sec = 3600` and `startup_timeout_sec = 120` (defaults are 60 s / 10 s — too short for a long video or the first model download) |
+| Google Antigravity | agent panel → ⋯ → *MCP Servers* → *Manage MCP Servers* → *View raw config*, add the JSON below |
+| Gemini CLI | add the JSON below to `~/.gemini/settings.json` |
+| Cursor | add the JSON below to `~/.cursor/mcp.json` |
+| VS Code (Copilot) | `.vscode/mcp.json`: same entry under `"servers"` with `"type": "stdio"` |
+```json
+{
+  "mcpServers": {
+    "babelscribe": { "command": "uvx", "args": ["babelscribe", "mcp"], "timeout": 3600000 }
+  }
+}
+```
+(`timeout` is in milliseconds and only some apps read it.) Already installed with pip? Use `"command": "babelscribe", "args": ["mcp"]`.
+Web-only chat apps (e.g. grok.com, chatgpt.com) can only reach servers on the internet, not your computer, so they can't use your GPU or files.
+## Benchmark: real talks, human captions as the answer key
+Four TED / TEDx talks, scored against the **human-made captions in the spoken language** (`bench/bench.py`, reproducible).
+Error = word error rate (WER) for space-separated languages, character error rate (CER) for Japanese and Thai.
+GPU: AMD Radeon RX 9070 XT via Vulkan. *default* = `large-v3-turbo`; *accurate* = `--accurate` (see the FLEURS section).
+| Language | Talk | Length | default: time / error | `--accurate`: time / error |
+|---|---|---|---|---|
+| English | [Matt Walker — Sleep Is Your Superpower (TED)](https://youtu.be/5MuIMqhT8DM) | 19.3 min | 48 s (24x) · WER **1.5%** | 107 s (11x) · WER **1.5%** |
+| Japanese | [Kazunari Taguchi (TEDxHimi)](https://youtu.be/cjtmDEG-B7U) | 16.2 min | 47 s (20x) · CER 4.7% | 115 s (8x) · CER **4.2%** |
+| Spanish | [Adrià Solà Pastor — Cómo hablar (TEDxESIC University)](https://youtu.be/XUqrvbsTfck) | 19.9 min | 56 s (21x) · WER 13.8% | 134 s (9x) · WER 13.5% |
+| Thai | [นิติ ชัยชิตาทร — โปรดเรียกฉันด้วยนามอันแท้จริง (TEDxBangkok)](https://youtu.be/48A9SU6_bQ8) | 14.1 min | 74 s (12x) · CER 22.7% | 565 s (2x) · CER **16.4%** |
+How to read it: TED captions are edited for reading (fillers dropped, light rewording), so these numbers are an upper bound —
+most of the Spanish "errors" are the speaker's actual words versus the tidied caption. Thai is genuinely harder: fast,
+casual speech with slang; `--accurate` (Pathumma Whisper text + turbo timing) cuts its error by more than a quarter.
+Talks are used only to measure accuracy; their transcripts are not redistributed (TED content is CC BY-NC-ND).
+## Benchmark: 16 languages on FLEURS
+[Google FLEURS](https://huggingface.co/datasets/google/fleurs) test set, 50 utterances per language, human-verified
+verbatim transcripts (`bench/fleurs.py`, reproducible). Same normaliser for every language (Whisper's rule: lower-case,
+drop punctuation and non-spacing marks, numbers spelled out). WER for space-separated languages, CER for ja / zh / ko / th.
+| Language | default (turbo) | `--accurate` | model `--accurate` uses | `--accurate`, edge-trimmed* |
+|---|---|---|---|---|
+| English | 6.1% | 5.7% | large-v3, beam 5 | 5.6% |
+| Spanish | 4.1% | 4.2% | large-v3, beam 5 | **2.6%** |
+| French | 7.4% | 7.3% | large-v3, beam 5 | 7.2% |
+| German | 4.3% | 4.0% | large-v3, beam 5 | **3.3%** |
+| Portuguese | 8.5% | 7.7% | large-v3, beam 5 | 5.3% |
+| Italian | 6.7% | 6.5% | large-v3, beam 5 | **3.8%** |
+| Russian | 6.8% | 5.8% | large-v3, beam 5 | 5.6% |
+| Arabic | 11.0% | 10.6% | large-v3, beam 5 | 9.7% |
+| Hindi | 28.4% | **12.6%** | [vasista22/whisper-hindi-large-v2](https://huggingface.co/vasista22/whisper-hindi-large-v2) + turbo timing | 11.1% |
+| Indonesian | 9.8% | 7.8% | large-v3, beam 5 | 5.6% |
+| Vietnamese | 10.8% | 9.1% | large-v3, beam 5 | 9.1% |
+| Turkish | 6.2% | 6.7% | large-v3, beam 5 | 5.8% |
+| Japanese (CER) | 6.4% | 5.7% | large-v3, beam 5 | **4.6%** |
+| Chinese (CER) | 6.3% | 5.3% | large-v3, beam 5 | **4.8%** |
+| Korean (CER) | 4.1% | 3.9% | large-v3, beam 5 | **3.1%** |
+| Thai (CER) | 15.9% | **8.9%** | [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) (NECTEC) + turbo timing | 8.8% |
+\* Some FLEURS clips contain more speech than their reference transcript, so a correct model is charged for words the
+reference leaves out. *Edge-trimmed* ignores extra words before the first / after the last reference word; both numbers
+are stored by `bench/fleurs.py`. 50 utterances per language means differences under ~0.5 points are noise.
+Also measured and **not** used, because large-v3 was as good or better: large-v2 (all languages),
+PhoWhisper-large (vi 19.0%), whisper-large-v3 dialectal / code-switching Arabic fine-tunes (12.2% / 15.2%),
+Typhoon Whisper (th 11.9%), Thonburian Whisper (th 9.1% — kept as an option), Vaani Hindi (16.5%).
+## Why babelscribe (vs. what already exists)
+| | GPU on AMD / Intel | Windows, no build step | Video in, subtitles out | Long files don't loop | Better text for your language |
+|---|---|---|---|---|---|
+| **babelscribe** | ✅ Vulkan | ✅ prebuilt `whisper-cli` downloaded for you | ✅ ffmpeg bundled | ✅ `--max-context 0` by default | ✅ hybrid fine-tune text + turbo timing |
+| whisper.cpp (raw) | ✅ Vulkan — if you compile it | ❌ official releases ship no Windows Vulkan build | ❌ WAV 16 kHz only | ⚠️ you must know the flag | ❌ |
+| faster-whisper / WhisperX | ❌ GPU = NVIDIA CUDA only | ✅ pip | ✅ | ⚠️ | ⚠️ manual |
+| Cloud APIs | n/a (cloud) | ✅ | ✅ | ✅ | ❌ — and your audio leaves your machine, paid per minute |
+babelscribe does **not** replace those projects — it stands on whisper.cpp and simply removes the hard parts:
+compiling for your GPU, converting media, picking the right device, avoiding the long-file repeat bug, and combining
+a language-specific fine-tune with accurate timestamps.
+## FAQ
+**How do I run Whisper on an AMD GPU on Windows?**
+`pip install babelscribe`, then `babelscribe video.mp4`. It downloads a Vulkan build of whisper.cpp that runs on AMD Radeon
+(and Intel Arc / NVIDIA) cards on Windows and Linux — no ROCm, no CUDA, no compiling.
+**How do I make subtitles (SRT) from a video for free, offline?**
+`babelscribe video.mp4 -f srt` writes `video.srt` next to the video. Nothing is uploaded; it runs on your own computer.
+**What is the most accurate free transcription for Thai?**
+`babelscribe video.mp4 -l th --accurate` — Pathumma Whisper (NECTEC) for the text plus Whisper turbo for timing:
+CER 8.9% on Google FLEURS vs 15.9% for plain Whisper turbo. Thonburian Whisper is available too (`--text-model thai-thonburian`).
+**Can Claude / ChatGPT Codex / Gemini transcribe a video on my computer?**
+Yes — add babelscribe as an MCP server (see *Use it from an AI app*). Claude Desktop installs it with one click from
+`babelscribe.mcpb`. The AI app can then find a file, transcribe it on your GPU, and proofread, translate or summarise the text.
+**How fast is it?**
+About 20x real time with Whisper large-v3-turbo on an AMD Radeon RX 9070 XT: a 19-minute talk in 48 seconds.
+**Which languages are supported?**
+All 99 Whisper languages, with automatic language detection. FLEURS error rates for 16 of them are in the benchmark table.
+**Is it better than faster-whisper or WhisperX?**
+Those are excellent on NVIDIA GPUs; on AMD / Intel GPUs they fall back to the CPU. babelscribe's niche is any GPU, zero setup,
+and better Thai / Hindi through community fine-tunes. If you have an NVIDIA card and like Python, faster-whisper is a fine choice.
+## Languages
+All 99 languages Whisper was trained on, auto-detected or forced with `-l`:
+af am ar as az ba be bg bn bo br bs ca cs cy da de el en es et eu fa fi fo fr gl gu ha haw he hi hr ht hu hy id is it ja jw ka kk km kn ko la lb ln lo lt lv mg mi mk ml mn mr ms mt my ne nl nn no oc pa pl ps pt ro ru sa sd si sk sl sn so sq sr su sv sw ta te tg th tk tl tr tt uk ur uz vi yi yo yue zh
+Accuracy follows Whisper's own training data: excellent for high-resource languages (English, Spanish, Japanese, …),
+weaker for low-resource ones. That is what hybrid mode is for — a community fine-tune for one language can be plugged in
+with one line in `babelscribe/models.py` (Thai and Hindi so far). PRs adding fine-tunes for other
+languages are the most valuable contribution.
+## Limitations (honest)
+- Tested end to end so far on an AMD Radeon RX 9070 XT (Windows, Vulkan). CUDA, Linux and macOS builds are produced by CI; reports from those machines are welcome.
+- Hybrid mode runs two models, so it is slower (≈1 min per minute of audio with a large fine-tune on that GPU).
+- Proper nouns can still be misspelled — check names before publishing subtitles.
+## Features
+- **Any input** — mp4, mkv, mov, mp3, wav, m4a… (ffmpeg is bundled through `imageio-ffmpeg`).
+- **Any language** — `-l auto` detects it; or pass `-l th`, `-l ja`, `-l es`…
+- **Picks the right GPU** — prefers a discrete card over an integrated one; `babelscribe devices` lists them, `--device N` overrides.
+- **Long files that don't loop** — runs whisper with `--max-context 0`, which stops the classic "same sentence repeated forever" hallucination on long recordings.
+- **Hybrid mode for better spelling in your language** — community fine-tunes (e.g. Thai *Thonburian Whisper*) spell far better but often lose timestamps. `--text-model` takes the text from the fine-tune and the timing from `large-v3-turbo`, aligned character by character (works for languages without spaces), cut only at word boundaries, with the timing model filling any words the fine-tune skipped.
+- **`--accurate`** — slower, fewest errors: large-v3 with beam search, or for Thai and Hindi the best community fine-tune (hybrid). Chosen per language from the FLEURS benchmark above.
+- **Outputs** — `srt`, `vtt`, `txt`, `json` (segments with token timings).
+## Usage
+```bash
+babelscribe talk.mp4 -f srt,vtt,txt,json          # all formats
+babelscribe podcast.mp3 -l en -m large-v3          # pick language and model
+babelscribe talk.mp4 -l hi --accurate              # slower, fewest errors (best model per language)
+babelscribe vo.wav -l th --text-model thai-thonburian   # hybrid: pick a Thai fine-tune yourself
+babelscribe devices                                # GPUs whisper.cpp can see
+babelscribe models                                 # models and fine-tunes
+babelscribe talk.mp4 --bin /path/to/whisper-cli    # use your own whisper.cpp build
+```
+From Python:
+```python
+from babelscribe.api import transcribe_file
+r = transcribe_file("talk.mp4", lang="auto", formats=["srt", "txt"])   # or accurate=True
+print(r["lang"], r["device"], r["files"]); print(r["text"][:200])
+```
+Models download on first use to `~/.babelscribe/models` (`BABELSCRIBE_MODELS` to change). Hybrid fine-tunes are converted on your
+machine from their original Hugging Face repo — install `pip install "babelscribe[finetune]"` once; converted weights are never
+redistributed, so each fine-tune keeps its own licence. Thai word boundaries: `pip install "babelscribe[thai]"`.
+## Add a language fine-tune
+Add one entry to `FINETUNES` in `babelscribe/models.py` (Hugging Face repo, language code, whether it needs beam search) and open a PR.
+## Prebuilt binaries
+`.github/workflows/build-binaries.yml` builds `whisper-cli` + `whisper-quantize` for every row of the table above and attaches
+them to each `v*` release. Point `BABELSCRIBE_RELEASES` at another URL to self-host.
+## ภาษาไทย
+ใช้ผ่าน Claude Desktop ได้โดยไม่ต้องพิมพ์คำสั่ง: โหลด `babelscribe.mcpb` จากหน้า Releases แล้วดับเบิลคลิก จากนั้นพิมพ์ในแชทว่า *"ทำซับไทยให้คลิปล่าสุดในโฟลเดอร์ดาวน์โหลด"*
+ถอดเสียงจากไฟล์เสียงหรือวิดีโอได้ทุกภาษา บนการ์ดจอทุกยี่ห้อ (AMD / NVIDIA / Intel ผ่าน Vulkan, Apple ผ่าน Metal) หรือ CPU
+ภาษาไทยแนะนำ `babelscribe ไฟล์.mp4 -l th --accurate` — ข้อความจาก Pathumma Whisper (NECTEC) ที่ผิดน้อยที่สุดใน FLEURS (CER 8.9% เทียบ turbo 15.9%) + เวลาจาก large-v3-turbo
+หรือเลือก Thonburian Whisper เอง: `--text-model thai-thonburian`
+## Privacy Policy
+babelscribe runs entirely on your own computer. Last updated 2026-10-06.
+- **Data collection:** none. babelscribe has no telemetry, analytics, accounts or crash reporting, and never uploads your
+  audio, video, transcripts or file names anywhere.
+- **What it processes and where:** the media file you choose is converted and transcribed locally; subtitles and text are
+  written to your disk (next to the file, or the folder you choose). Through MCP, the transcript text is returned to the AI
+  app that called the tool — what that app does with it is governed by that app's own privacy policy.
+- **Network access (downloads only):** on first use it downloads the `whisper-cli` program from this project's GitHub
+  releases, speech models from Hugging Face (`huggingface.co/ggerganov/whisper.cpp`, and for `--accurate` Thai/Hindi the
+  fine-tune's own repository), and Python packages from PyPI when installed with pip/uv. These are plain downloads;
+  no personal data is sent. GitHub, Hugging Face and PyPI see a normal download request (IP address, user agent) under
+  their own privacy policies.
+- **Storage and retention:** downloaded programs and models are cached in `~/.babelscribe` (or `BABELSCRIBE_HOME` /
+  `BABELSCRIBE_MODELS`) until you delete that folder. Outputs stay wherever they were written until you delete them.
+  babelscribe keeps no other data.
+- **Third-party sharing:** none.
+- **Contact:** open an issue at https://github.com/phonology024/babelscribe/issues
+## Credits & licence
+MIT. Built on [whisper.cpp](https://github.com/ggml-org/whisper.cpp) (MIT) and OpenAI Whisper models (MIT).
+Fine-tunes belong to their authors: [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) by NECTEC,
+[Thonburian Whisper](https://huggingface.co/biodatlab/whisper-th-large-v3-combined) by biodatlab,
+[whisper-hindi-large-v2](https://huggingface.co/vasista22/whisper-hindi-large-v2) by vasista22 (Speech Lab, IIT Madras).

{babelscribe-0.3.1 → babelscribe-0.3.2}/babelscribe/__init__.py RENAMED Viewed

@@ -1,2 +1,2 @@
 """babelscribe — any language, any GPU speech-to-text built on whisper.cpp."""
-__version__ = "0.3.1"
+__version__ = "0.3.2"

{babelscribe-0.3.1 → babelscribe-0.3.2}/babelscribe/mcp_server.py RENAMED Viewed

@@ -1,136 +1,139 @@
-"""babelscribe as an MCP server: any MCP client (Claude Desktop / Code, Codex, Antigravity, Gemini CLI, Cursor,
-VS Code, ...) can transcribe local audio/video on the user's own GPU.
-  babelscribe mcp            # stdio server; AI apps launch it themselves (see README "Use it from an AI app")
-Needs Python 3.10+ (the mcp SDK)."""
-import io
-import os
-import sys
-import time
-from pathlib import Path
-MEDIA = {".mp4", ".mkv", ".mov", ".avi", ".webm", ".m4v", ".mp3", ".wav", ".m4a", ".aac", ".flac", ".ogg", ".opus", ".wma"}
-TEXT_LIMIT = 30000   # characters of transcript returned to the model; the full text is always in the files
-def _guard_stdout():
-    """stdout is the JSON-RPC channel. Keep a private handle to it for the protocol and point everything else
-    (print, logging, stray library output) at stderr so it can never corrupt a message."""
-    proto = io.TextIOWrapper(os.fdopen(os.dup(sys.stdout.fileno()), "wb"), encoding="utf-8", newline="\n",
-                             write_through=True)
-    sys.stdout = sys.stderr
-    return proto
-def build():
-    import anyio
-    from mcp.server.mcpserver import Context, MCPServer
-    from . import __version__, api, backend, models
-    srv = MCPServer("babelscribe", version=__version__, instructions=(
-        "Local speech-to-text on the user's own GPU (AMD/NVIDIA/Intel/Apple) for 99 languages. Needs a file path on "
-        "this computer: if the user only names or describes a file, call find_media first. Subtitles (.srt) and text "
-        "(.txt) are written next to the media file unless output_dir is given. Use accurate=true when the user wants "
-        "the fewest mistakes (slower); it is much better for Thai and Hindi. The first run downloads the model "
-        "(~1.6 GB), so it can take a few minutes. After transcribing you can proofread names, translate, or "
-        "summarise from the returned text."))
-    @srv.tool()
-    async def transcribe(file_path: str, language: str = "auto", accurate: bool = False, formats: str = "srt,txt",
-                         output_dir: str = "", ctx: Context | None = None) -> dict:
-        """Transcribe a local audio or video file into subtitles / text.
-        file_path: absolute path to the media file (mp4, mkv, mov, mp3, wav, m4a, ...).
-        language: ISO code such as en, th, ja, es, hi — or "auto" to detect.
-        accurate: slower, fewest errors (large-v3 with beam search, or the best fine-tune for Thai / Hindi).
-        formats: comma list from srt, vtt, txt, json.
-        output_dir: folder for the output files; empty = next to the media file.
-        Returns the files written, detected language, device used, and the transcript text."""
-        src = Path(file_path).expanduser()
-        out = Path(output_dir).expanduser() / src.stem if output_dir else None
-        if out:
-            out.parent.mkdir(parents=True, exist_ok=True)
-        fmts = [f.strip() for f in formats.split(",") if f.strip()]
-        lines: list[str] = []
-        done = anyio.Event()
-        async def heartbeat():   # progress keeps clients from timing out on long files
-            t0 = time.time()
-            while not done.is_set():
-                with anyio.move_on_after(5):
-                    await done.wait()
-                if ctx and not done.is_set():
-                    await ctx.report_progress(time.time() - t0, None, lines[-1] if lines else "starting ...")
-        result: dict = {}
-        async with anyio.create_task_group() as tg:
-            tg.start_soon(heartbeat)
-            try:
-                result = await anyio.to_thread.run_sync(lambda: api.transcribe_file(
-                    src, language, accurate=accurate, formats=fmts, out=out, log=lambda m: (lines.append(m), print(m))))
-            finally:
-                done.set()
-        text = result.pop("text", "")
-        result["text"] = text[:TEXT_LIMIT]
-        if len(text) > TEXT_LIMIT:
-            result["note"] = f"transcript truncated to {TEXT_LIMIT} characters; the full text is in the files listed"
-        return result
-    @srv.tool()
-    def find_media(name_contains: str = "", folder: str = "", limit: int = 15) -> list[dict]:
-        """Find audio/video files on this computer, newest first. Searches Downloads, Videos, Desktop, Music and
-        Documents (two levels deep) unless folder is given. Use it when the user names a file without a full path."""
-        home = Path.home()
-        roots = [Path(folder).expanduser()] if folder else [home / d for d in ("Downloads", "Videos", "Desktop", "Music", "Documents", "Movies")]
-        hits = []
-        for r in roots:
-            if not r.is_dir():
-                continue
-            for p in [*r.glob("*"), *r.glob("*/*"), *(r.glob("*/*/*") if folder else [])]:
-                if p.suffix.lower() in MEDIA and name_contains.lower() in p.name.lower():
-                    try:
-                        st = p.stat()
-                    except OSError:
-                        continue
-                    hits.append((st.st_mtime, p, st.st_size))
-        hits.sort(reverse=True)
-        return [{"path": str(p), "size_mb": round(sz / 1e6, 1), "modified": time.strftime("%Y-%m-%d %H:%M", time.localtime(t))}
-                for t, p, sz in hits[:limit]]
-    @srv.tool()
-    def list_devices() -> dict:
-        """GPUs whisper.cpp can use on this computer and which one babelscribe picks automatically."""
-        _, _, found = api.setup()
-        return {"devices": found, "auto_pick": backend.pick_device(found)}
-    @srv.tool()
-    def list_languages_and_models() -> dict:
-        """Models available, and which model --accurate uses per language."""
-        return {"general": models.GENERAL, "fine_tunes": {k: v["note"] for k, v in models.FINETUNES.items()},
-                "accurate_per_language": {**models.ACCURATE, "*": "large-v3 beam 5"}}
-    return srv
-def main() -> None:
-    try:
-        import anyio
-        from mcp.server.stdio import stdio_server
-    except ImportError:
-        raise SystemExit('the MCP server needs Python 3.10+ (the mcp SDK)')
-    proto = _guard_stdout()
-    srv = build()
-    async def run():
-        async with stdio_server(stdout=anyio.wrap_file(proto)) as (r, w):
-            low = srv._lowlevel_server
-            await low.run(r, w, low.create_initialization_options())
-    anyio.run(run)
-if __name__ == "__main__":
-    main()
+"""babelscribe as an MCP server: any MCP client (Claude Desktop / Code, Codex, Antigravity, Gemini CLI, Cursor,
+VS Code, ...) can transcribe local audio/video on the user's own GPU.
+  babelscribe mcp            # stdio server; AI apps launch it themselves (see README "Use it from an AI app")
+Needs Python 3.10+ (the mcp SDK)."""
+import io
+import os
+import sys
+import time
+from pathlib import Path
+MEDIA = {".mp4", ".mkv", ".mov", ".avi", ".webm", ".m4v", ".mp3", ".wav", ".m4a", ".aac", ".flac", ".ogg", ".opus", ".wma"}
+TEXT_LIMIT = 30000   # characters of transcript returned to the model; the full text is always in the files
+def _guard_stdout():
+    """stdout is the JSON-RPC channel. Keep a private handle to it for the protocol and point everything else
+    (print, logging, stray library output) at stderr so it can never corrupt a message."""
+    proto = io.TextIOWrapper(os.fdopen(os.dup(sys.stdout.fileno()), "wb"), encoding="utf-8", newline="\n",
+                             write_through=True)
+    sys.stdout = sys.stderr
+    return proto
+def build():
+    import anyio
+    from mcp.server.mcpserver import Context, MCPServer
+    from mcp.types import ToolAnnotations
+    from . import __version__, api, backend, models
+    srv = MCPServer("babelscribe", version=__version__, instructions=(
+        "Local speech-to-text on the user's own GPU (AMD/NVIDIA/Intel/Apple) for 99 languages. Needs a file path on "
+        "this computer: if the user only names or describes a file, call find_media first. Subtitles (.srt) and text "
+        "(.txt) are written next to the media file unless output_dir is given. Use accurate=true when the user wants "
+        "the fewest mistakes (slower); it is much better for Thai and Hindi. The first run downloads the model "
+        "(~1.6 GB), so it can take a few minutes. After transcribing you can proofread names, translate, or "
+        "summarise from the returned text."))
+    # transcribe writes new .srt/.txt files (re-running rewrites its own output); first use downloads whisper-cli and the model
+    @srv.tool(title="Transcribe audio/video to subtitles", annotations=ToolAnnotations(
+        read_only_hint=False, destructive_hint=False, idempotent_hint=True, open_world_hint=True))
+    async def transcribe(file_path: str, language: str = "auto", accurate: bool = False, formats: str = "srt,txt",
+                         output_dir: str = "", ctx: Context | None = None) -> dict:
+        """Transcribe a local audio or video file into subtitles / text.
+        file_path: absolute path to the media file (mp4, mkv, mov, mp3, wav, m4a, ...).
+        language: ISO code such as en, th, ja, es, hi — or "auto" to detect.
+        accurate: slower, fewest errors (large-v3 with beam search, or the best fine-tune for Thai / Hindi).
+        formats: comma list from srt, vtt, txt, json.
+        output_dir: folder for the output files; empty = next to the media file.
+        Returns the files written, detected language, device used, and the transcript text."""
+        src = Path(file_path).expanduser()
+        out = Path(output_dir).expanduser() / src.stem if output_dir else None
+        if out:
+            out.parent.mkdir(parents=True, exist_ok=True)
+        fmts = [f.strip() for f in formats.split(",") if f.strip()]
+        lines: list[str] = []
+        done = anyio.Event()
+        async def heartbeat():   # progress keeps clients from timing out on long files
+            t0 = time.time()
+            while not done.is_set():
+                with anyio.move_on_after(5):
+                    await done.wait()
+                if ctx and not done.is_set():
+                    await ctx.report_progress(time.time() - t0, None, lines[-1] if lines else "starting ...")
+        result: dict = {}
+        async with anyio.create_task_group() as tg:
+            tg.start_soon(heartbeat)
+            try:
+                result = await anyio.to_thread.run_sync(lambda: api.transcribe_file(
+                    src, language, accurate=accurate, formats=fmts, out=out, log=lambda m: (lines.append(m), print(m))))
+            finally:
+                done.set()
+        text = result.pop("text", "")
+        result["text"] = text[:TEXT_LIMIT]
+        if len(text) > TEXT_LIMIT:
+            result["note"] = f"transcript truncated to {TEXT_LIMIT} characters; the full text is in the files listed"
+        return result
+    @srv.tool(title="Find audio/video files", annotations=ToolAnnotations(read_only_hint=True, open_world_hint=False))
+    def find_media(name_contains: str = "", folder: str = "", limit: int = 15) -> list[dict]:
+        """Find audio/video files on this computer, newest first. Searches Downloads, Videos, Desktop, Music and
+        Documents (two levels deep) unless folder is given. Use it when the user names a file without a full path."""
+        home = Path.home()
+        roots = [Path(folder).expanduser()] if folder else [home / d for d in ("Downloads", "Videos", "Desktop", "Music", "Documents", "Movies")]
+        hits = []
+        for r in roots:
+            if not r.is_dir():
+                continue
+            for p in [*r.glob("*"), *r.glob("*/*"), *(r.glob("*/*/*") if folder else [])]:
+                if p.suffix.lower() in MEDIA and name_contains.lower() in p.name.lower():
+                    try:
+                        st = p.stat()
+                    except OSError:
+                        continue
+                    hits.append((st.st_mtime, p, st.st_size))
+        hits.sort(reverse=True)
+        return [{"path": str(p), "size_mb": round(sz / 1e6, 1), "modified": time.strftime("%Y-%m-%d %H:%M", time.localtime(t))}
+                for t, p, sz in hits[:limit]]
+    @srv.tool(title="List GPUs", annotations=ToolAnnotations(read_only_hint=True, open_world_hint=True))
+    def list_devices() -> dict:
+        """GPUs whisper.cpp can use on this computer and which one babelscribe picks automatically."""
+        _, _, found = api.setup()
+        return {"devices": found, "auto_pick": backend.pick_device(found)}
+    @srv.tool(title="List languages and models", annotations=ToolAnnotations(read_only_hint=True, open_world_hint=False))
+    def list_languages_and_models() -> dict:
+        """Models available, and which model --accurate uses per language."""
+        return {"general": models.GENERAL, "fine_tunes": {k: v["note"] for k, v in models.FINETUNES.items()},
+                "accurate_per_language": {**models.ACCURATE, "*": "large-v3 beam 5"}}
+    return srv
+def main() -> None:
+    try:
+        import anyio
+        from mcp.server.stdio import stdio_server
+    except ImportError:
+        raise SystemExit('the MCP server needs Python 3.10+ (the mcp SDK)')
+    proto = _guard_stdout()
+    srv = build()
+    async def run():
+        async with stdio_server(stdout=anyio.wrap_file(proto)) as (r, w):
+            low = srv._lowlevel_server
+            await low.run(r, w, low.create_initialization_options())
+    anyio.run(run)
+if __name__ == "__main__":
+    main()

{babelscribe-0.3.1 → babelscribe-0.3.2}/babelscribe.egg-info/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: babelscribe
-Version: 0.3.1
+Version: 0.3.2
 Summary: Transcribe any audio/video in 99 languages on any GPU (AMD, NVIDIA, Intel via Vulkan; NVIDIA via CUDA; Apple via Metal) or CPU — whisper.cpp with batteries included.
 License: MIT
 Project-URL: Homepage, https://github.com/phonology024/babelscribe
@@ -251,6 +251,25 @@ them to each `v*` release. Point `BABELSCRIBE_RELEASES` at another URL to self-h
 ภาษาไทยแนะนำ `babelscribe ไฟล์.mp4 -l th --accurate` — ข้อความจาก Pathumma Whisper (NECTEC) ที่ผิดน้อยที่สุดใน FLEURS (CER 8.9% เทียบ turbo 15.9%) + เวลาจาก large-v3-turbo
 หรือเลือก Thonburian Whisper เอง: `--text-model thai-thonburian`
+## Privacy Policy
+babelscribe runs entirely on your own computer. Last updated 2026-10-06.
+- **Data collection:** none. babelscribe has no telemetry, analytics, accounts or crash reporting, and never uploads your
+  audio, video, transcripts or file names anywhere.
+- **What it processes and where:** the media file you choose is converted and transcribed locally; subtitles and text are
+  written to your disk (next to the file, or the folder you choose). Through MCP, the transcript text is returned to the AI
+  app that called the tool — what that app does with it is governed by that app's own privacy policy.
+- **Network access (downloads only):** on first use it downloads the `whisper-cli` program from this project's GitHub
+  releases, speech models from Hugging Face (`huggingface.co/ggerganov/whisper.cpp`, and for `--accurate` Thai/Hindi the
+  fine-tune's own repository), and Python packages from PyPI when installed with pip/uv. These are plain downloads;
+  no personal data is sent. GitHub, Hugging Face and PyPI see a normal download request (IP address, user agent) under
+  their own privacy policies.
+- **Storage and retention:** downloaded programs and models are cached in `~/.babelscribe` (or `BABELSCRIBE_HOME` /
+  `BABELSCRIBE_MODELS`) until you delete that folder. Outputs stay wherever they were written until you delete them.
+  babelscribe keeps no other data.
+- **Third-party sharing:** none.
+- **Contact:** open an issue at https://github.com/phonology024/babelscribe/issues
 ## Credits & licence
 MIT. Built on [whisper.cpp](https://github.com/ggml-org/whisper.cpp) (MIT) and OpenAI Whisper models (MIT).
 Fine-tunes belong to their authors: [Pathumma Whisper](https://huggingface.co/nectec/Pathumma-whisper-th-large-v3) by NECTEC,

{babelscribe-0.3.1 → babelscribe-0.3.2}/pyproject.toml RENAMED Viewed

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "babelscribe"
-version = "0.3.1"
+version = "0.3.2"
 description = "Transcribe any audio/video in 99 languages on any GPU (AMD, NVIDIA, Intel via Vulkan; NVIDIA via CUDA; Apple via Metal) or CPU — whisper.cpp with batteries included."
 readme = "README.md"
 license = { text = "MIT" }