Merge branch 'bugfix/vad_cache' into 'master'

Bugfix/vad cache

See merge request speech-recognition-framework/esp-sr!185

.gitlab-ci.yml

@@ -143,7 +143,7 @@ build_esp_sr:
     - .rules:build:test_esp_sr
   parallel:
     matrix:
-      - IMAGE: [espressif/idf:v5.4.1, espressif/idf:latest]
+      - IMAGE: [espressif/idf:release-v5.4, espressif/idf:latest]
         EXAMPLES_PATH: "test_apps/esp-sr"
 
 
@@ -153,7 +153,7 @@ build_esp_tts:
     - .rules:build:test_esp_tts
   parallel:
     matrix:
-      - IMAGE: [espressif/idf:v5.4.1, espressif/idf:latest]
+      - IMAGE: [espressif/idf:release-v5.4, espressif/idf:latest]
         EXAMPLES_PATH: "test_apps/esp-tts"
 
 build_esp32c5:
@@ -162,7 +162,7 @@ build_esp32c5:
     - .rules:build:test_esp32c5
   parallel:
     matrix:
-      - IMAGE: [espressif/idf:v5.4.1, espressif/idf:latest]
+      - IMAGE: [espressif/idf:release-v5.4, espressif/idf:latest]
         EXAMPLES_PATH: "test_apps/esp32c5"
 
 .test_template: &test_template

CHANGELOG.md

@@ -1,5 +1,9 @@
 # Change log for esp-sr
 
+## 2.1.5
+- Fix ringbuf bug
+- Add more wakenet9 models
+
 ## 2.1.4
 - Fix the bug of multinet duration overflow
 - Fix the memory leak bug in multinet7 and multinet6

Kconfig.projbuild

@@ -112,10 +112,18 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "Hey,Willow (wn9_heywillow_tts)"
     default False
 
+    config SR_WN_WN9_HIMFIVE
+    bool "Hi,M Five (wn9_himfive)"
+    default False
+
     config SR_WN_WN9_SOPHIA_TTS
     bool "Sophia (wn9_sophia_tts)"
     default False
 
+    config SR_WN_WN9_HEYWANDA_TTS
+    bool "Hey,Wanda (wn9_heywanda_tts)"
+    default False
+
     config SR_WN_WN9_HIJOLLY_TTS2
     bool "Hi,Jolly (wn9_hijolly_tts2)"
     default False
@@ -124,36 +132,10 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "Hi,Fairy (wn9_hifairy_tts2)"
     default False
 
-    config SR_WN_WN9_NIHAOXIAOXIN_TTS
-    bool "你好小鑫 (wn9_nihaoxiaoxin_tts)"
-    default False
-
-    config SR_WN_WN9_XIAOMEITONGXUE_TTS
-    bool "小美同学 (wn9_xiaomeitongxue_tts)"
-    default False
-
     config SR_WN_WN9_HEYPRINTER_TTS
     bool "Hey,Printer (wn9_heyprinter_tts)"
     default False
 
-    config SR_WN_WN9_XIAOLONGXIAOLONG_TTS
-    bool "小龙小龙 (wn9_xiaolongxiaolong_tts)"
-    default False
-
-    config SR_WN_WN9_MIAOMIAOTONGXUE_TTS
-    bool "喵喵同学 (wn9_miaomiaotongxue_tts)"
-    default False
-
-
-    config SR_WN_WN9_HEYWANDA_TTS
-    bool "Hey,Wanda (wn9_heywanda_tts)"
-    default False
-
-    config SR_WN_WN9_HIMIAOMIAO_TTS
-    bool "Hi,喵喵 (wn9_himiaomiao_tts)"
-    default False
-
-
     config SR_WN_WN9_MYCROFT_TTS
     bool "Mycroft (wn9_mycroft_tts)"
     default False
@@ -162,6 +144,42 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "Hi,Joy (wn9_hijoy_tts)"
     default False
 
+    config SR_WN_WN9_HIJASON_TTS2
+    bool "Hi,Jason (wn9_hijason_tts2)"
+    default False
+
+    config SR_WN_WN9_ASTROLABE_TTS
+    bool "Astrolabe (wn9_astrolabe_tts)"
+    default False
+
+    config SR_WN_WN9_HEYILY_TTS2
+    bool "Hey,Ily (wn9_heyily_tts2)"
+    default False
+
+    config SR_WN_WN9_BLUECHIP_TTS2
+    bool "Blue Chip (wn9_bluechip_tts2)"
+    default False
+
+    config SR_WN_WN9_XIAOLONGXIAOLONG_TTS
+    bool "小龙小龙 (wn9_xiaolongxiaolong_tts)"
+    default False
+
+    config SR_WN_WN9_HIMIAOMIAO_TTS
+    bool "Hi,喵喵 (wn9_himiaomiao_tts)"
+    default False
+
+    config SR_WN_WN9_MIAOMIAOTONGXUE_TTS
+    bool "喵喵同学 (wn9_miaomiaotongxue_tts)"
+    default False
+
+    config SR_WN_WN9_NIHAOXIAOXIN_TTS
+    bool "你好小鑫 (wn9_nihaoxiaoxin_tts)"
+    default False
+
+    config SR_WN_WN9_XIAOMEITONGXUE_TTS
+    bool "小美同学 (wn9_xiaomeitongxue_tts)"
+    default False
+
     config SR_WN_WN9_HILILI_TTS
     bool "Hi,Lily/Hi,莉莉 (wn9_hilili_tts)"
     default False
@@ -178,22 +196,10 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "Hi,小巫 (wn9_haixiaowu_tts)"
     default False
 
-    config SR_WN_WN9_ASTROLABE_TTS
-    bool "Astrolabe (wn9_astrolabe_tts)"
-    default False
-
-    config SR_WN_WN9_HEYILY_TTS2
-    bool "Hey,Ily (wn9_heyily_tts2)"
-    default False
-
     config SR_WN_WN9_XIAOYAXIAOYA_TTS2
     bool "小鸭小鸭 (wn9_xiaoyaxiaoya_tts2)"
     default False
 
-    config SR_WN_WN9_HIJASON_TTS2
-    bool "Hi,Jason (wn9_hijason_tts2)"
-    default False
-
     config SR_WN_WN9_LINAIBAN_TTS2
     bool "璃奈板 (wn9_linaiban_tts2)"
     default False
@@ -226,7 +232,6 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "你好小益 (wn9_nihaoxiaoyi_tts2)"
     default False
 
-
     config SR_WN_WN9_NIHAOBAIYING_TTS2
     bool "你好百应 (wn9_nihaobaiying_tts2)"
     default False
@@ -239,14 +244,13 @@ menu "Load Multiple Wake Words (WakeNet9)"
     bool "Hi Wall E or Hi 瓦力(wn9_hiwalle_tts2)"
     default False
 
-    config SR_WN_WN9_HIMFIVE
-    bool "Hi,M Five (wn9_himfive)"
-    default False
-
-
     config SR_WN_WN9_XIAOLUXIAOLU_TTS2
     bool "小鹿小鹿 (wn9_xiaoluxiaolu_tts2)"
     default False
+
+    config SR_WN_WN9_NIHAOXIAOAN_TTS2
+    bool "你好小安 (wn9_nihaoxiaoan_tts2)"
+    default False
 endmenu
 
 

README.md

@@ -60,6 +60,7 @@ The following wake words are supported in esp-sr:
 |Hey,Ily          |                                  | wn9_heyily_tts2        |
 |Hi,Jolly         |                                  | wn9_hijolly_tts2        |
 |Hi,Fairy         |                                  | wn9_hifairy_tts2        |
+|Blue Chip        |                                  | wn9_bluechip_tts2        |
 |Hi,Wall E/Hi,瓦力|                                  | wn9_hiwalle_tts2       |
 |你好小鑫         |                                  | wn9_nihaoxiaoxin_tts   |
 |小美同学         |                                  | wn9_xiaomeitongxue_tts |
@@ -83,6 +84,7 @@ The following wake words are supported in esp-sr:
 |你好百应         |                                  | wn9_nihaobaiying_tts2|
 |小鹿小鹿         |                                  | wn9_xiaoluxiaolu_tts2|
 |你好东东         |                                  | wn9_nihaodongdong_tts2|
+|你好小安         |                                  | wn9_nihaoxiaoan_tts2|
 
 *NOTE:* `_tts` suffix means this WakeNet model is trained by TTS samples. `_tts2` suffix means this WakeNet model is trained by TTS Pipeline V2.
 

idf_component.yml

@@ -1,4 +1,4 @@
-version: "2.1.4"
+version: "2.1.5"
 description: esp_sr provides basic algorithms for Speech Recognition applications
 url: https://github.com/espressif/esp-sr
 dependencies:

include/esp32/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32c3/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32c5/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32c6/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32p4/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32s2/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

include/esp32s3/esp_mn_iface.h

@@ -47,7 +47,8 @@ typedef struct{
     int command_id[ESP_MN_RESULT_MAX_NUM];     // The list of command id.
     int phrase_id[ESP_MN_RESULT_MAX_NUM];      // The list of phrase id.
     float prob[ESP_MN_RESULT_MAX_NUM];         // The list of probability.
-    char string[256];
+    char string[256];                          // recognized string with commands graph
+    char raw_string[256];                      // recognized string without commands graph
 } esp_mn_results_t;
 
 typedef struct {

model/wakenet_model/wn9_bluechip_tts2/_MODEL_INFO_

@@ -0,0 +1 @@
+wakenet9_tts2h12_BlueChip_3_0.622_0.626

model/wakenet_model/wn9_nihaoxiaoan_tts2/_MODEL_INFO_

@@ -0,0 +1 @@
+wakenet9_tts2h12_nihaoxiaoan_3_0.621_0.625