Upload README.md with huggingface_hub

README.md

@@ -1,11 +1,10 @@
-# Supertonic – FP16 vs INT8 Quantized Benchmark (by Shadow0482)
+# Supertonic – FP16 vs INT8 Quantized Benchmark (Shadow0482)
 
 This README documents a simple benchmark comparing **FP16** and **INT8** quantized
 versions of the [Supertonic](https://huggingface.co/Supertone/supertonic) TTS
 pipeline using the quantized models hosted at:
 
 - Quantized models repo: **Shadow0482/supertonic-quantized**
-- This model card: **Shadow0482/supertonic-quantized**
 
 All tests were run in Google Colab on CPU using the official `py/example_onnx.py`
 script from the Supertonic GitHub repository.
@@ -22,11 +21,10 @@ The same text was used for both FP16 and INT8 runs:
 
 ## Results
 
-| Variant | Precision | ONNX directory        | Time (s) | Output WAV                      | Status |
-|--------:|-----------|-----------------------|---------:|----------------------------------|--------|
-| FP16    | float16   | `onnx_fp16/`          | 0.289 | `NONE` | FAILED |
-| INT8    | int8      | `onnx_int8/`          | 0.248 | `NONE` | FAILED |
-
+| Variant | Precision | ONNX directory | Time (s) | Output WAV | Status |
+|--------:|-----------|----------------|---------:|-----------:|--------|
+| FP16    | float16   | `onnx_fp16/`   | 0.914 | `NONE` | FAILED |
+| INT8    | int8      | `onnx_int8/`   | 6.644 | `Greetings__You_are_l_1.wav` | OK |
 
 
 > Note:
@@ -36,45 +34,33 @@ The same text was used for both FP16 and INT8 runs:
 
 ---
 
-## How to reproduce this benchmark
-
-In a fresh Colab notebook:
+## How this benchmark was run
 
-1. Install dependencies and run the benchmark cell (the one that created this README).
-2. Make sure you have write access to both:
-   - `Shadow0482/supertonic`
-   - `Shadow0482/supertonic-quantized`
-
-The core of the benchmark is simply:
+1. Clone the official Supertonic repository and download its assets (configs + voice styles).
+2. Download `Shadow0482/supertonic-quantized` and copy:
+   - `fp16/*.fp16.onnx` → `onnx_fp16/*.onnx`
+   - `int8_dynamic/*.int8.onnx` → `onnx_int8/*.onnx`
+3. Copy configuration files:
+   - `assets/configs/*.json`
+   - `assets/onnx/tts.json`, `assets/onnx/unicode_indexer.json`
+4. Run:
 
 ```bash
 python py/example_onnx.py \
   --onnx-dir onnx_fp16 \
   --voice-style assets/voice_styles/M1.json \
-  --text "... your test text ..." \
+  --text "..." \
   --n-test 1 \
   --save-dir results_fp16
 
 python py/example_onnx.py \
   --onnx-dir onnx_int8 \
   --voice-style assets/voice_styles/M1.json \
-  --text "... the same test text ..." \
+  --text "..." \
   --n-test 1 \
   --save-dir results_int8
 ````
 
-Where `onnx_fp16/` and `onnx_int8/` contain drop-in copies of the original
-Supertonic ONNX files, but converted/quantized to FP16 or INT8 respectively.
-
----
-
-## Model notes
-
-* **FP16** models are converted from the original FP32 weights using float16 conversion.
-* **INT8** models are dynamically quantized (MatMul/Gemm) using ONNX Runtime.
-* The quantized models live in `{QUANT_REPO_ID}` and can be plugged into the
-  Supertonic pipeline via the `--onnx-dir` argument in `example_onnx.py`.
-
 ---
 
 ## License