From 79013d80b5073b004c154fa70546320f86566723 Mon Sep 17 00:00:00 2001
From: yanko <yanko@idrobots.com>
Date: Mon, 16 Feb 2026 19:08:18 +0200
Subject: [PATCH 1/2] Fix voice installer: split pip commands and pin
 Pillow>=10

pip 22's resolver won't downgrade huggingface-hub (installed by
faster-whisper) to satisfy transformers<5 in a single command, so it
silently picks transformers 5.x. Split into two pip calls so the
downgrade happens reliably.

Also pin Pillow>=10 because Ubuntu 22.04's system Pillow 9.0.1 is
missing PIL.Image.Resampling, which transformers 4.x needs at import.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 scripts/install-voice.sh | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/scripts/install-voice.sh b/scripts/install-voice.sh
index dd7e331d0..e08c12365 100755
--- a/scripts/install-voice.sh
+++ b/scripts/install-voice.sh
@@ -89,7 +89,11 @@ fi
 # ── Step 4: Install Kokoro TTS ───────────────────────────────────────────────
 
 echo "[4/7] Installing Kokoro TTS..."
-su - "$CLAWBOX_USER" -c "$PIP install --user 'numpy<2' 'transformers<5' kokoro soundfile Pillow" 2>&1 | tail -3
+# Install kokoro first, then force transformers<5 as a separate step.
+# pip 22's resolver won't downgrade huggingface-hub (pulled in by faster-whisper)
+# to satisfy transformers<5 in a single command, so it silently picks transformers 5.x.
+su - "$CLAWBOX_USER" -c "$PIP install --user 'numpy<2' kokoro soundfile 'Pillow>=10'" 2>&1 | tail -3
+su - "$CLAWBOX_USER" -c "$PIP install --user 'transformers<5'" 2>&1 | tail -3
 
 # ── Step 5: Pre-download models ─────────────────────────────────────────────
 

From 128da7093bc9bda38f55a721177eed78897e788e Mon Sep 17 00:00:00 2001
From: yanko <yanko@idrobots.com>
Date: Mon, 16 Feb 2026 19:21:07 +0200
Subject: [PATCH 2/2] Add cmake to voice deps and clear corrupt Whisper cache
 on install
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CTranslate2 CUDA build requires cmake and build-essential — add them
to the apt-get in step_voice_install().

Also detect and clear 0-byte HuggingFace cache blobs before the
Whisper model download step, preventing "failed to read a value"
errors from corrupted/incomplete prior downloads.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 install.sh               | 2 +-
 scripts/install-voice.sh | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/install.sh b/install.sh
index 07be64269..ca7de7387 100755
--- a/install.sh
+++ b/install.sh
@@ -322,7 +322,7 @@ step_polkit_rules() {
 }
 
 step_voice_install() {
-  apt-get install -y -qq espeak-ng libsndfile1
+  apt-get install -y -qq espeak-ng libsndfile1 cmake build-essential
   bash "$PROJECT_DIR/scripts/install-voice.sh"
   echo "  Voice pipeline installed"
   bash "$PROJECT_DIR/scripts/setup-optimizations.sh"
diff --git a/scripts/install-voice.sh b/scripts/install-voice.sh
index e08c12365..64387aa4d 100755
--- a/scripts/install-voice.sh
+++ b/scripts/install-voice.sh
@@ -98,6 +98,12 @@ su - "$CLAWBOX_USER" -c "$PIP install --user 'transformers<5'" 2>&1 | tail -3
 # ── Step 5: Pre-download models ─────────────────────────────────────────────
 
 echo "[5/7] Pre-downloading Whisper model (base)..."
+# Clear corrupted cache (0-byte blobs from failed/rate-limited HF downloads)
+WHISPER_CACHE="$CLAWBOX_HOME/.cache/huggingface/hub/models--Systran--faster-whisper-base"
+if [ -d "$WHISPER_CACHE/blobs" ] && find "$WHISPER_CACHE/blobs" -maxdepth 1 -type f -empty | grep -q .; then
+  echo "  Clearing corrupted Whisper model cache..."
+  rm -rf "$WHISPER_CACHE"
+fi
 DEVICE="cpu"
 COMPUTE="auto"
 if $HAS_CUDA; then