doc: Update mn6 docs

docs/en/benchmark/README.rst

@@ -151,7 +151,7 @@ Resource Occupancy
     | MultiNet 5  | 16 KB       | 2310 KB     | 12 ms       | 32 ms       |
     | Q8          |             |             |             |             |
     +-------------+-------------+-------------+-------------+-------------+
-    | MultiNet 6  | 52 KB       | 4400 KB     | 12 ms       | 32 ms       |
+    | MultiNet 6  | 48 KB       | 4000 KB     | 12 ms       | 32 ms       |
     +-------------+-------------+-------------+-------------+-------------+
 
 Performance Test

docs/en/speech_command_recognition/README.rst

@@ -23,17 +23,7 @@ MultiNet is a lightweight model designed to recognize multiple speech command wo
 
 The MultiNet input is the audio processed by the audio-front-end algorithm (AFE), with the format of 16 KHz, 16 bit and mono. By recognizing the audio signals, speech commands can be recognized.
 
-The following table shows the models supported by Espressif SoCs:
-
-+---------+-----------+-------------+---------------+-------------+
-|  Chip   |   ESP32   |   ESP32S3                                 |
-+=========+===========+=============+===============+=============+
-| Model   | MultiNet2 | MultiNet4.5 | MultiNet4.5Q8 | MultiNet5Q8 |
-+---------+-----------+-------------+---------------+-------------+
-| Chinese | √         | √           | √             | √           |
-+---------+-----------+-------------+---------------+-------------+
-| English |           |             |               | √           |
-+---------+-----------+-------------+---------------+-------------+
+Please refer to :doc:`Models Benchmark  <../benchmark/README>` to check models supported by Espressif SoCs.
 
 For details on flash models, see Section :doc:`Flashing Models <../flash_model/README>` .
 

docs/zh_CN/benchmark/README.rst

@@ -151,7 +151,7 @@ MultiNet
     | MultiNet 5  | 16 KB       | 2310 KB     | 12 ms       | 32 ms       |
     | Q8          |             |             |             |             |
     +-------------+-------------+-------------+-------------+-------------+
-    | MultiNet 6  | 52 KB       | 4400 KB     | 12 ms       | 32 ms       |
+    | MultiNet 6  | 48 KB       | 4000 KB     | 12 ms       | 32 ms       |
     +-------------+-------------+-------------+-------------+-------------+
 
 性能测试

docs/zh_CN/speech_command_recognition/README.rst

@@ -23,17 +23,7 @@ MultiNet 是为了在 {IDF_TARGET_NAME} 系列上离线实现多命令词识别
 
 MultiNet 输入为经过前端语音算法（AFE）处理过的音频（格式为 16 KHz，16 bit，单声道）。通过对音频进行识别，则可以对应到相应的汉字或单词。
 
-以下表格展示在不同芯片上的模型支持：
-
-+---------+-----------+-------------+---------------+-------------+
-|  Chip   |   ESP32   |   ESP32S3                                 |
-+=========+===========+=============+===============+=============+
-| Model   | MultiNet2 | MultiNet4.5 | MultiNet4.5Q8 | MultiNet5Q8 |
-+---------+-----------+-------------+---------------+-------------+
-| Chinese | √         | √           | √             | √           |
-+---------+-----------+-------------+---------------+-------------+
-| English |           |             |               | √           |
-+---------+-----------+-------------+---------------+-------------+
+请参考 :doc:`Models Benchmark  <../benchmark/README>` 去查看当前不同芯片支持的模型。
 
 用户选择不同的模型的方法请参考 :doc:`模型加载 <../flash_model/README>` 。
 

tool/README.md

@@ -1,3 +1,26 @@
+## MultiNet6
+
+#### Step 1. Data preparation
+
+For English, words are used as units. Please prepare a list of commands written in a text file `commands_en.txt` of the following format:
+
+```
+# command_id command_sentence
+1 TELL ME A JOKE
+2 MAKE A COFFEE
+```
+
+For Chinese, pinyin are used as units. Please prepare a list of commands written in a text file `commands_cn.txt` of the following format:
+```
+# command_id command_sentence
+1 da kai kong tiao
+2 guan bi kong tiao
+```
+
+#### Step 2. Move created files
+
+1. Move your `commands_en.txt` or `commands_cn.txt` to `/model/multinet_model/fst/`
+
 ## MultiNet5
 ####  1. Install g2p_en, please refer to https://pypi.org/project/g2p-en/ 
 
@@ -42,45 +65,3 @@ multinet->reset(model_data, new_commands_str, err_id);
 // turn off the light -> commond id=2
 ```
 
-## MultiNet6
-
-
-The FST (Finite State Transducer) is used to save a list of commands.
-
-#### Step 1. Data preparation
-
-Requirements:
-- python>3.8
-- sentencepiece
-
-To create a FST from a list of commands, two files are needed:
-- commands.txt: maps a command id to subwords
-- tokens.txt: maps subword tokens to it's indices in the bpe model
-
-Assume you have a list of commands written in a text file `commands_list.txt` of the following format:
-
-```
-# command_id command_sentence
-1 TELL ME A JOKE
-2 MAKE A COFFEE
-```
-**Note**: command ids starts from 1, 0 is reserved in FST.
-
-Run the following command to create the required files, do not change the filenames `commands.txt` and `tokens.txt`. 
-
-```sh
-pip install -r requirements.txt
-
-python fst/prepare_for_fst.py \
-    --infile commands_list.txt \
-    --bpe-model fst/bpe.model \
-    --out-command-list commands.txt \
-    --out-token-symbols tokens.txt
-```
-
-#### Step 2. Move created files
-
-1. Remove `/model/multinet_model/fst/fst.txt` and `/model/multinet_model/fst/fst_reversed.txt` if those files exist. 
-2. Move the following files to `/model/multinet_model/fst/`
-- commands.txt
-- tokens.txt