README: v1.3.5

README.md

@@ -50,22 +50,21 @@ Download **one** build and run the installer:
 | Build | For | Installer |
 |---|---|---|
 | **GPU (CUDA)** | NVIDIA GPU (Compute Capability ≥ 7.0) — fastest; bundles the CUDA 13.2 + cuDNN 9 runtime | [**MedpodInstaller-1.3.0-cuda.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.0-cuda.exe) |
-| **CPU** | Any 64-bit Windows PC with **AVX2** (Intel Haswell 2013+ / AMD 2015+) | [**MedpodInstaller-1.3.4-cpu.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.4-cpu.exe) |
+| **CPU** | Any 64-bit Windows PC with **AVX2** (Intel Haswell 2013+ / AMD 2015+) | [**MedpodInstaller-1.3.5-cpu.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.5-cpu.exe) |
 
 The installers are **code-signed** (Microsoft Azure Trusted Signing) and
 self-contained — the GPU build bundles the NVIDIA CUDA/cuDNN runtime, so there is
 nothing else to install for the app itself.
 
-> ⚡ **What's new in v1.3.4 (CPU builds).** **Long-audio transcription quality fix.**
-> MedAudio now splits long recordings into large segments cut only at natural pauses
-> (never mid-word) — benchmarked to match whole-clip accuracy while transcribing
-> ~3× faster on a 10-minute clip. The per-tab performance footers were also refreshed
-> (MedVision shows image-**embedding** speed; MedAudio drops the not-applicable
-> prefill figure). Earlier 1.3.x wins remain: faster prompt processing on by default,
-> an optimized quantized kernel + correct multi-core scheduling, generation speed
-> **flat over long replies**, ~3× faster time-to-first-token, lower memory use
-> (half-precision KV cache), and a live **tok/s readout** under each chat reply. Just
-> download the new CPU installer.
+> ⚡ **What's new in v1.3.5 (CPU builds).** **Safer recording controls + clearer
+> versioning.** MedAudio's buttons now guide you through Record → Pause/Continue →
+> Finish, only enable **Transcribe** once a recording is finished, and turn a stopped
+> recording's cancel into **Delete Recording** with a confirm prompt — so you can't
+> accidentally wipe a recording; recording is clearly flagged in **red**. The app now
+> shows its **real version and build** (CPU/GPU + CPU variant) on the sign-in screen
+> and in Settings. Image-analysis reports also stream more smoothly. Previous 1.3.4
+> long-audio transcription quality fix and all earlier 1.3.x chat speedups remain.
+> Just download the new CPU installer.
 
 **Optimized CPU builds (advanced).** Two extra CPU builds give faster prompt
 processing and image analysis on newer CPUs. **Each requires the matching CPU
@@ -74,8 +73,8 @@ build above (it runs everywhere).
 
 | Optimized build | Best for | Requires |
 |---|---|---|
-| [**MedpodInstaller-1.3.4-cpu-raptorlake.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.4-cpu-raptorlake.exe) | Intel 13th/14th-gen Core (i5/i7/i9) — adds **AVX-VNNI** | Intel 12th-gen+ Core (AVX-VNNI) |
-| [**MedpodInstaller-1.3.4-cpu-avx512.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.4-cpu-avx512.exe) | AMD Ryzen 7000/9000 (Zen 4/5) and AVX-512 Intel (Ice Lake, Xeon, 11th-gen desktop) — adds **AVX-512** | Any CPU with AVX-512 |
+| [**MedpodInstaller-1.3.5-cpu-raptorlake.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.5-cpu-raptorlake.exe) | Intel 13th/14th-gen Core (i5/i7/i9) — adds **AVX-VNNI** | Intel 12th-gen+ Core (AVX-VNNI) |
+| [**MedpodInstaller-1.3.5-cpu-avx512.exe**](https://huggingface.co/cloudkites/medpodgp/resolve/main/MedpodInstaller-1.3.5-cpu-avx512.exe) | AMD Ryzen 7000/9000 (Zen 4/5) and AVX-512 Intel (Ice Lake, Xeon, 11th-gen desktop) — adds **AVX-512** | Any CPU with AVX-512 |
 
 ### Step 2 — Install the AI models