Add ability to record Raven examples in web UI

TODO.md

@@ -21,6 +21,7 @@
 * Raven
     * Compute MFCC for new frame only
     * Use C/C++ for extra speed
+    * Output WAV data with detection
 
 # TTS
 

VERSION

@@ -1 +1 @@
-2.5.3
+2.5.4

docs/wake-word.md

@@ -29,7 +29,7 @@ Rhasspy listens for `hermes/hotword/<wakewordId>/detected` messages to decide wh
 
 ## Raven
 
-Listens for a wake word with [raven](https://github.com/rhasspy/rhasspy-wake-raven). This system is based on [Snips Personal Wakeword Detector](https://medium.com/snips-ai/machine-learning-on-voice-a-gentle-introduction-with-snips-personal-wake-word-detector-133bd6fb568e).
+Listens for a wake word with [Raven](https://github.com/rhasspy/rhasspy-wake-raven). This system is based on the [Snips Personal Wakeword Detector](https://medium.com/snips-ai/machine-learning-on-voice-a-gentle-introduction-with-snips-personal-wake-word-detector-133bd6fb568e) and works by comparing incoming audio to several pre-recorded templates.
 
 Add to your [profile](profiles.md):
 
@@ -43,7 +43,7 @@ Add to your [profile](profiles.md):
 }
 ```
 
-You will need to record at least 3 WAV template files with your custom wake word. Once recorded, trim silence from the audio using a program like [Audacity](https://www.audacityteam.org/). Export the templates to a directory named `raven` in your profile as 16-bit 16Khz mono WAV files.
+To train Raven, you will need to record at least 3 WAV template files with your custom wake word. This can be done in the Rhasspy web interface or manually with a program like [Audacity](https://www.audacityteam.org/). If you record manually, make sure to trim silence from the beginning and end of the audio and export the templates to a directory named `raven` in your profile as 16-bit 16Khz mono WAV files.
 
 You can adjust the sensitivty by changing `raven.probability_threshold` to a be a tighter (less sensitive) or looser (more sensitive) range around 0.5. Additionally, you can increase the value of `minmum_matches` to required more than one WAV template to match before a detection occurs. This should reduce false positives, but may increase false negatives.