#picovoice — Public Fediverse posts

Coffee by Command: The Speech2Touch Voice Hack - If you were to troll your colleagues, you can label your office coffee maker any d... - https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #peripheralshacks #softwarehacks #cookinghacks #frankea600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack - If you were to troll your colleagues, you can label your office coffee maker any d... - https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #peripheralshacks #softwarehacks #cookinghacks #frankea600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack - If you were to troll your colleagues, you can label your office coffee maker any d... - https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #peripheralshacks #softwarehacks #cookinghacks #frankea600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack - If you were to troll your colleagues, you can label your office coffee maker any d... - https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #peripheralshacks #softwarehacks #cookinghacks #frankea600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack - If you were to troll your colleagues, you can label your office coffee maker any d... - https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #peripheralshacks #softwarehacks #cookinghacks #frankea600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #PeripheralsHacks #SoftwareHacks #cookinghacks #FrankeA600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #PeripheralsHacks #SoftwareHacks #cookinghacks #FrankeA600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #PeripheralsHacks #SoftwareHacks #cookinghacks #FrankeA600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

Coffee by Command: The Speech2Touch Voice Hack https://hackaday.com/2025/09/24/coffee-by-command-the-speech2touch-voice-hack/ #PeripheralsHacks #SoftwareHacks #cookinghacks #FrankeA600 #homehacks #picovoice #coffee #speech #stm32 #voice #hid #usb

The upcoming version of #Platypush will include a revised assistant.picovoice integration as the advised #voice assistant plugin - assistant.google is based on a library deprecated long ago and it’s becoming harder and harder to make it work on newer systems.

I’m overall very impressed by the state of the #Picovoice products - much better than the last time I checked them out a few years ago. It’s now possible to build a full-featured voice assistant using their hotword+STT+TTS engines, and I’m very impressed by the flexibility of their intent detection too.

Platypush will now allow setting up your custom ChatGPT-powered voice assistant in something as simple as:

from platypush import hook, run
from platypush.message.event.assistant import (
    SpeechRecognizedEvent,
    ResponseEndEvent,
)

@hook(SpeechRecognizedEvent)
def on_speech_detected(event, **_):
    response = run(
        "openai.get_response", prompt=event.phrase
    )

    event.assistant.render_response(response)

@hook(ResponseEndEvent)
def on_response_end(event, **_):
    # If the AI response is a question, automatically start
    # a follow-up interaction
    if event.response_text and \
         event.response_text.endswith("?"):
        event.assistant.start_conversation()

The adventurous among you can already start playing with the new integration by checking out the latest master branch.

In the meantime, I’m packing the documentation and preparing a new blog post about it, while preparing the new release, so stay tuned!

https://git.platypush.tech/platypush/platypush/src/commit/72bc6971222af8cef51046e89e83106fac0e3dc0/platypush/plugins/assistant/picovoice/__init__.py#L14

The upcoming version of #Platypush will include a revised assistant.picovoice integration as the advised #voice assistant plugin - assistant.google is based on a library deprecated long ago and it’s becoming harder and harder to make it work on newer systems.

I’m overall very impressed by the state of the #Picovoice products - much better than the last time I checked them out a few years ago. It’s now possible to build a full-featured voice assistant using their hotword+STT+TTS engines, and I’m very impressed by the flexibility of their intent detection too.

Platypush will now allow setting up your custom ChatGPT-powered voice assistant in something as simple as:

from platypush import hook, run
from platypush.message.event.assistant import (
    SpeechRecognizedEvent,
    ResponseEndEvent,
)

@hook(SpeechRecognizedEvent)
def on_speech_detected(event, **_):
    response = run(
        "openai.get_response", prompt=event.phrase
    )

    event.assistant.render_response(response)

@hook(ResponseEndEvent)
def on_response_end(event, **_):
    # If the AI response is a question, automatically start
    # a follow-up interaction
    if event.response_text and \
         event.response_text.endswith("?"):
        event.assistant.start_conversation()

The adventurous among you can already start playing with the new integration by checking out the latest master branch.

In the meantime, I’m packing the documentation and preparing a new blog post about it, while preparing the new release, so stay tuned!

https://git.platypush.tech/platypush/platypush/src/commit/72bc6971222af8cef51046e89e83106fac0e3dc0/platypush/plugins/assistant/picovoice/__init__.py#L14

The upcoming version of #Platypush will include a revised assistant.picovoice integration as the advised #voice assistant plugin - assistant.google is based on a library deprecated long ago and it’s becoming harder and harder to make it work on newer systems.

I’m overall very impressed by the state of the #Picovoice products - much better than the last time I checked them out a few years ago. It’s now possible to build a full-featured voice assistant using their hotword+STT+TTS engines, and I’m very impressed by the flexibility of their intent detection too.

Platypush will now allow setting up your custom ChatGPT-powered voice assistant in something as simple as:

from platypush import hook, run
from platypush.message.event.assistant import (
    SpeechRecognizedEvent,
    ResponseEndEvent,
)

@hook(SpeechRecognizedEvent)
def on_speech_detected(event, **_):
    response = run(
        "openai.get_response", prompt=event.phrase
    )

    event.assistant.render_response(response)

@hook(ResponseEndEvent)
def on_response_end(event, **_):
    # If the AI response is a question, automatically start
    # a follow-up interaction
    if event.response_text and \
         event.response_text.endswith("?"):
        event.assistant.start_conversation()

The adventurous among you can already start playing with the new integration by checking out the latest master branch.

In the meantime, I’m packing the documentation and preparing a new blog post about it, while preparing the new release, so stay tuned!

https://git.platypush.tech/platypush/platypush/src/commit/72bc6971222af8cef51046e89e83106fac0e3dc0/platypush/plugins/assistant/picovoice/__init__.py#L14

The upcoming version of #Platypush will include a revised assistant.picovoice integration as the advised #voice assistant plugin - assistant.google is based on a library deprecated long ago and it’s becoming harder and harder to make it work on newer systems.

I’m overall very impressed by the state of the #Picovoice products - much better than the last time I checked them out a few years ago. It’s now possible to build a full-featured voice assistant using their hotword+STT+TTS engines, and I’m very impressed by the flexibility of their intent detection too.

Platypush will now allow setting up your custom ChatGPT-powered voice assistant in something as simple as:

from platypush import hook, run
from platypush.message.event.assistant import (
    SpeechRecognizedEvent,
    ResponseEndEvent,
)

@hook(SpeechRecognizedEvent)
def on_speech_detected(event, **_):
    response = run(
        "openai.get_response", prompt=event.phrase
    )

    event.assistant.render_response(response)

@hook(ResponseEndEvent)
def on_response_end(event, **_):
    # If the AI response is a question, automatically start
    # a follow-up interaction
    if event.response_text and \
         event.response_text.endswith("?"):
        event.assistant.start_conversation()

The adventurous among you can already start playing with the new integration by checking out the latest master branch.

In the meantime, I’m packing the documentation and preparing a new blog post about it, while preparing the new release, so stay tuned!

https://git.platypush.tech/platypush/platypush/src/commit/72bc6971222af8cef51046e89e83106fac0e3dc0/platypush/plugins/assistant/picovoice/__init__.py#L14

Update on the state of #voice integrations in #Platypush: my initial despair (Snowboy and Mycroft are dead, Mozilla DeepSpeech is abandoned, OpenAI’s Whisper is very resource-hungry, and the Google Assistant library is deprecated and still alive just because I’m keeping those bindings alive in Platypush) turned into excitement after I’ve played a bit more with #Picovoice.

RE: https://manganiello.blog/api/posts/yp1zetjnrw

Update on the state of #voice integrations in #Platypush: my initial despair (Snowboy and Mycroft are dead, Mozilla DeepSpeech is abandoned, OpenAI’s Whisper is very resource-hungry, and the Google Assistant library is deprecated and still alive just because I’m keeping those bindings alive in Platypush) turned into excitement after I’ve played a bit more with #Picovoice.

RE: https://manganiello.blog/api/posts/yp1zetjnrw

Update on the state of #voice integrations in #Platypush: my initial despair (Snowboy and Mycroft are dead, Mozilla DeepSpeech is abandoned, OpenAI’s Whisper is very resource-hungry, and the Google Assistant library is deprecated and still alive just because I’m keeping those bindings alive in Platypush) turned into excitement after I’ve played a bit more with #Picovoice.

RE: https://manganiello.blog/api/posts/yp1zetjnrw

Update on the state of #voice integrations in #Platypush: my initial despair (Snowboy and Mycroft are dead, Mozilla DeepSpeech is abandoned, OpenAI’s Whisper is very resource-hungry, and the Google Assistant library is deprecated and still alive just because I’m keeping those bindings alive in Platypush) turned into excitement after I’ve played a bit more with #Picovoice.

RE: https://manganiello.blog/api/posts/yp1zetjnrw

Since #OpenAI released their ChatGPT Turbo & Whisper API's a few days ago, there's loads of voice > Whisper STT > ChatGPT Turbo API > TTS demos, so I had to do one..

Though this uses Wake Word and Voice Activity Detection using
#PicoVoice so it's kinda like an Alexa.. ish.

Google's TTS is pretty slow though. Any way to speed it up?

Since #OpenAI released their ChatGPT Turbo & Whisper API's a few days ago, there's loads of voice > Whisper STT > ChatGPT Turbo API > TTS demos, so I had to do one..

Though this uses Wake Word and Voice Activity Detection using
#PicoVoice so it's kinda like an Alexa.. ish.

Google's TTS is pretty slow though. Any way to speed it up?

Since #OpenAI released their ChatGPT Turbo & Whisper API's a few days ago, there's loads of voice > Whisper STT > ChatGPT Turbo API > TTS demos, so I had to do one..

Though this uses Wake Word and Voice Activity Detection using
#PicoVoice so it's kinda like an Alexa.. ish.

Google's TTS is pretty slow though. Any way to speed it up?

Since #OpenAI released their ChatGPT Turbo & Whisper API's a few days ago, there's loads of voice > Whisper STT > ChatGPT Turbo API > TTS demos, so I had to do one..

Though this uses Wake Word and Voice Activity Detection using
#PicoVoice so it's kinda like an Alexa.. ish.

Google's TTS is pretty slow though. Any way to speed it up?

Since #OpenAI released their ChatGPT Turbo & Whisper API's a few days ago, there's loads of voice > Whisper STT > ChatGPT Turbo API > TTS demos, so I had to do one..

Though this uses Wake Word and Voice Activity Detection using
#PicoVoice so it's kinda like an Alexa.. ish.

Google's TTS is pretty slow though. Any way to speed it up?

#Tips: If you want to use a custom wake-word for SEPIA #OpenAssistant created with Porcupine (2.1) by #Picovoice you can now use this website to convert your '.ppn' file to the right JS format: https://sepia-framework.github.io/files/porcupine/convert.html (the old b64 files are not included anymore in the v2.1 Zip file)

#Tips: If you want to use a custom wake-word for SEPIA #OpenAssistant created with Porcupine (2.1) by #Picovoice you can now use this website to convert your '.ppn' file to the right JS format: https://sepia-framework.github.io/files/porcupine/convert.html (the old b64 files are not included anymore in the v2.1 Zip file)

#Tips: If you want to use a custom wake-word for SEPIA #OpenAssistant created with Porcupine (2.1) by #Picovoice you can now use this website to convert your '.ppn' file to the right JS format: https://sepia-framework.github.io/files/porcupine/convert.html (the old b64 files are not included anymore in the v2.1 Zip file)

je cherche un APK #android pour faire du #Text-to-Speech en français et depuis n'importe que champ de saisie sans envoyer mes requêtes chez quelqu'un.

Je ne trouve pas de binaire pour #Mozilla #deepspeech #picovoice ou #vosk.

Vous avez des recommandations?

je cherche un APK #android pour faire du #Text-to-Speech en français et depuis n'importe que champ de saisie sans envoyer mes requêtes chez quelqu'un.

Je ne trouve pas de binaire pour #Mozilla #deepspeech #picovoice ou #vosk.

Vous avez des recommandations?

je cherche un APK #android pour faire du #Text-to-Speech en français et depuis n'importe que champ de saisie sans envoyer mes requêtes chez quelqu'un.

Je ne trouve pas de binaire pour #Mozilla #deepspeech #picovoice ou #vosk.

Vous avez des recommandations?