Use chatterbox TTS

1 files changed, 19 insertions, 35 deletions

diff --git a/makima/sh/download-models.sh b/makima/sh/download-models.sh
index 4f188f3..b44e091 100755
--- a/makima/sh/download-models.sh
+++ b/makima/sh/download-models.sh
@@ -114,47 +114,31 @@ else
     echo "All models downloaded successfully"
 fi
 
-# Download Qwen3-TTS models (for TTS functionality)
-QWEN3_TTS_DIR="${QWEN3_TTS_DIR:-/app/models/qwen3-tts}"
-
-download_qwen3_tts() {
-    if [ -d "$QWEN3_TTS_DIR" ] && \
-       [ -f "$QWEN3_TTS_DIR/model.safetensors" ] && \
-       [ -f "$QWEN3_TTS_DIR/speech_tokenizer.safetensors" ] && \
-       [ -f "$QWEN3_TTS_DIR/vocab.json" ] && \
-       [ -f "$QWEN3_TTS_DIR/merges.txt" ] && \
-       [ -f "$QWEN3_TTS_DIR/config.json" ]; then
-        echo "Qwen3-TTS models already exist, skipping..."
+# Download Chatterbox TTS models (for TTS functionality)
+CHATTERBOX_MODEL_DIR="${CHATTERBOX_MODEL_DIR:-/app/models/chatterbox-turbo}"
+
+download_chatterbox_tts() {
+    if [ -d "$CHATTERBOX_MODEL_DIR" ] && \
+       [ -f "$CHATTERBOX_MODEL_DIR/speech_encoder.onnx" ] && \
+       [ -f "$CHATTERBOX_MODEL_DIR/embed_tokens.onnx" ] && \
+       [ -f "$CHATTERBOX_MODEL_DIR/language_model.onnx" ] && \
+       [ -f "$CHATTERBOX_MODEL_DIR/conditional_decoder.onnx" ] && \
+       [ -f "$CHATTERBOX_MODEL_DIR/tokenizer.json" ]; then
+        echo "Chatterbox TTS models already exist, skipping..."
         return 0
     fi
 
-    echo "Downloading Qwen3-TTS models..."
-    mkdir -p "$QWEN3_TTS_DIR"
-
-    # Download base TTS model files from Qwen/Qwen3-TTS-12Hz-0.6B-Base
-    # Note: This repo uses vocab.json + merges.txt (not tokenizer.json)
-    echo "Downloading Qwen3-TTS-12Hz-0.6B-Base..."
-    hf download Qwen/Qwen3-TTS-12Hz-0.6B-Base \
-        model.safetensors \
-        config.json \
-        vocab.json \
-        merges.txt \
-        tokenizer_config.json \
-        --local-dir "$QWEN3_TTS_DIR"
-
-    # Download speech tokenizer from Qwen/Qwen3-TTS-Tokenizer-12Hz
-    echo "Downloading Qwen3-TTS-Tokenizer-12Hz..."
-    local tmpdir=$(mktemp -d)
-    hf download Qwen/Qwen3-TTS-Tokenizer-12Hz \
-        model.safetensors \
-        --local-dir "$tmpdir"
-    mv "$tmpdir/model.safetensors" "$QWEN3_TTS_DIR/speech_tokenizer.safetensors"
-    rm -rf "$tmpdir"
+    echo "Downloading Chatterbox TTS models..."
+    mkdir -p "$CHATTERBOX_MODEL_DIR"
+
+    # Download from ResembleAI/chatterbox-turbo-ONNX
+    echo "Downloading ResembleAI/chatterbox-turbo-ONNX..."
+    hf download ResembleAI/chatterbox-turbo-ONNX --local-dir "$CHATTERBOX_MODEL_DIR"
 
-    echo "Qwen3-TTS models downloaded successfully"
+    echo "Chatterbox TTS models downloaded successfully"
 }
 
-download_qwen3_tts
+download_chatterbox_tts
 
 # Execute the main command
 exec "$@"