Update 0.1B ONNX runtime usage guide

README.md

@@ -49,8 +49,10 @@ tokenizer, the packaged reference voice, and the optional FP16 encoder used to
 register reference voices.
 
 > **Model files only.** Inference, streaming service, and voice-registration
-> code live in the
-> [Audio8 TTS repository](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime).
+> code live in the dedicated
+> [0.1B INT8 ONNX Runtime](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime_0_1b_int8).
+> The older `onnx_runtime/` directory targets the incompatible 0.6B INT4
+> graphs and must not be used with this model.
 
 > **Acknowledgments**
 >
@@ -132,19 +134,21 @@ conditioned on the slow hidden state and preceding codebooks.
 
 ## Quick Start
 
-Python 3.11 or newer is required. The current release is tested on macOS
-arm64 and Linux x86_64.
+Python 3.10 or newer is required. The runtime uses ONNX Runtime's
+`CPUExecutionProvider`; platform-specific ONNX Runtime wheels may be needed on
+devices such as Jetson. The current release is tested on macOS arm64 and Linux
+x86_64.
 
 ### 1. Download the code and model
 
 ```bash
 git clone https://github.com/Audio8-AI/Audio8_TTS.git
-cd Audio8_TTS/onnx_runtime
+cd Audio8_TTS/onnx_runtime_0_1b_int8
 
 python3 -m pip install -U "huggingface_hub[cli]"
-hf download Audio8/Audio8-TTS-Preview-0.1B-ONNX-INT8 --local-dir model
+hf download Audio8/audio8-TTS-0.1B-ONNX-INT8 --local-dir model
 bash setup.sh
-python3 scripts/register_default_voice.py   # packaged fixed reference -> "default" voice
+"$PWD/.venv/bin/python" scripts/register_default_voice.py   # packaged reference -> "default" voice
 ```
 
 The model files are stored at this Hugging Face repository's root. Downloading
@@ -170,7 +174,6 @@ bash run_infer.sh \
   --voice default \
   --text "这是一个零样本语音合成的中文测试，声音克隆效果怎么样？" \
   --max-new-tokens 256 \
-  --greedy \
   --output outputs/example.wav
 ```
 
@@ -179,8 +182,11 @@ The command writes `outputs/example.wav` and `[10, T]` codec codes to
 
 ### 3. Register a reference voice
 
-Start the local service and open <http://127.0.0.1:8024>. Upload a 0.5-30
-second reference recording, its exact transcript, and a voice name.
+Start the local service and open <http://127.0.0.1:8024>. The page provides text
+generation, voice selection, WAV playback/download, reference-voice
+registration, memory status, and runtime reload. The Swagger API documentation
+is available at <http://127.0.0.1:8024/docs>. Upload a 0.5-30 second reference
+recording, its exact transcript, and a voice name to register a new voice.
 
 ```bash
 bash start_server.sh
@@ -215,7 +221,7 @@ curl http://127.0.0.1:8024/v1/audio/speech \
 ```
 
 See the complete
-[ONNX Runtime guide](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime)
+[0.1B ONNX Runtime guide](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime_0_1b_int8)
 for streaming output, configuration, memory management, and service controls.
 
 ## Evaluation
@@ -227,9 +233,9 @@ details.
 
 INT8 quantization can change sampled token sequences, so quality should be
 evaluated for each target language, voice, and deployment setting rather than
-assuming bit-for-bit equivalence with the source checkpoint. For
-bit-for-bit-faithful decoding, the repository also ships FP32 per-token graphs
-(see `onnx_runtime`).
+assuming bit-for-bit equivalence with the source checkpoint. This repository
+contains the 0.1B INT8 graphs; use the dedicated runtime linked above when
+running them.
 
 ## Limitations and Responsible Use
 
@@ -252,4 +258,3 @@ attribution details.
 
 We thank the Fish Audio team for publishing the DualAR architecture used in
 Fish Audio S2 Pro.
-