61

This is the main project my Raspberry Pi is dedicated to right now, so I figure I can add my two cents. Keep in mind this project is still very much a work in progress. I chose to use the C programming language for this project exclusively on the Raspbian OS, and that may have affected some of my decisions and instructions. I'm going to only list free and ...


6

Not possible with Google directly, but you can get other voice services installed. Here are 3 that I've tried on Raspberry Pi: Mycroft Open source, very capable https://github.com/MycroftAI/enclosure-picroft Alexa Very capable but you're locked down to using Amazon services Official: https://github.com/alexa/alexa-avs-sample-app Unofficial (but has a ...


6

This is not possible as of January 2017. This article from CNet explains it clearly: Here are eight things Amazon's assistant can do that Google's can't. [...] Back in March, Amazon released an API for Alexa Voice Service (AVS), the service which powers the Amazon Echo, Echo Dot and Amazon Tap. This allowed developers and users to put Alexa on ...


5

I went with pocketsphinx_continuous and a $4 sound card. To manage the fact that it needs to stop listening when using speech synth I used amixer to handle to input volume to the mic (this was recommended best practice by CMU as stop-starting engine will result in poorer recognition) echo "SETTING MIC IN TO 15 (94%)" >> ./audio.log amixer -c 1 set ...


4

I'd go so far to say that the computational power of an arduino would be quite a limit to voice recognition. Bitvoicer for example uses the arduino to sample the speech but not to process and recognice it. It sends the stream to a more powerful system to do the recognition part. The Pi on the other hand is powerful enough to do all the processing itself as ...


4

I think your best bet is to sit back and actually do more research. It seems that you don't quite know what you want to do, or don't know how to ask. This isn't a bad thing, it just seems that you are a little confused. Instead of just diving deep right into a larger project, maybe making smaller projects and moving on is a better bet? While NLP and AI are ...


4

I believe those (u092e, etc) are Hindi Unicode characters. Whatever software you're using to convert those characters into speech apparently doesn't handle them. You need to check its documentation relating to non-English Unicode to see if it can be enabled, or switch to another package that supports Hindi text-to-speech. UPDATE With the additional ...


4

Your tag says python3. If you are running your code in python3 you need to install speechrecognition with pip3.


3

W: Failed to fetch http://packages.cloud.google.com/apt/dists/cloud-sdk-jessie/InRelease If you go to that page you'll notice near the top: Architectures: amd64 i386 These refer to what are commonly known as x86-64 and x86 (32-bit) architectures; the vast majority of desktops and laptops now fall into one of these categories. The Raspberry Pi does not; ...


3

You are going to need an internet connection, an OS, (I suggest Raspbian) and a microphone. The first step is to get the microphone set-up. Start by opening a terminal window and running lusb. This lists all of the USB devices connected to the Pi and the microphone should show up. Next the volume for the mic needs to be set to high, so run alsamixer. You ...


3

I think the "nano" project - or parts thereof - (that soundly beat me in the Christmas Prizes for Best project of 2015) - might be of interest. It does feature both text-to-speech and speech-to-text if I recall correctly - and Patrick Cook is a currently active member here. Once your reputation gets to 20 (I think) you might catch him in The Bakery (the ...


3

Configure looks for the presence of the development libraries of PulseAudio, Jack and ALSA, in that order. That is why in answers like this it is mentioned that "...It also appears that ALSA will not be used if PulseAudio is installed...". In my particular case, I didn't have anything else installed besides ALSA. But then I remembered that I did some tests ...


3

Yes. use PocketSphinx for speech recognition, Festvox for text to speech (TTS) and some USB audio with line in (or an old supported webcam which also has line in). Google searches for these software packages and "Raspberry Pi" provide many examples and tutorials to set this up.


3

SiriProxy - Only use this if you have a device that uses Siri - you don't need to jailbreak anything. It basically intercepts Siri on the network you install it on. Speech2Text - You can use Googles API to decode speech to text but the example contains some other methods too. Julius - A speech recognition decoder. As pointed out by Lenik, you will need ...


2

Raspberry Pi has no built-in ADC nor microphone input. Unless you're planning to use external USB mike, there's basically no way to get your audio stream to the device. Besides that, there are no serious limitations, the CPU is powerful enough for any sound processing you might try to implement.


2

layla writes on http://robot.laylamah.com/?p=35 Note, if you receive an error such as the following: Error opening audio device plughw:1,0 for capture: Connection refused Mixer load failed: Invalid argument FATAL_ERROR: "continuous.c", line 246: Failed to open audio device You likely have pulseaudio installed, which is causing sphinxbase to attempt to use ...


2

Well I just find out that google speech API is not working as it used to anymore, now it ask for a captcha making it impossible to use for jasper, Now im using Wit.io and Festival


2

Yes there is Pocketsphinx to convert speech to text, that has python APIs For installation do: apt-cache search pocketsphinx and install all of them. Or I have a shorcut to do this: :D apt-get install $(apt-cache search pocketsphinx | \ awk '{print $1}' | tr '[:space:]' ' ') enjoy.. :)


1

Even the small custom dictionary was giving me the same error. Maybe this has to do something with the pyAudio itself. What helped in this case was setting "exception_on_overflow = False" as the stream method's second argument. This way at least the program continues to run. Let's see if it's enough to make the functionality work. p = pyaudio.PyAudio() ...


1

It is not possible to recognize large vocabulary speech on Raspberry Pi, it is too slow for that. This was discussed before. You can stream the data to the server or configure small grammar for recognition if you still want to recognize on the device. If you want to use smaller model, you still need to configure the sample rate since you record at sample ...


1

assistant pi is the way to go. installs both alexa and google home with hotword recognition - https://github.com/xtools-at/AssistantPi


1

Luckily things change. Google assistant integration was announced today 4/27/2017 http://www.theverge.com/2017/4/27/15442994/google-assistant-sdk-smart-speaker-actions-gadgets-raspberry-pi


1

According to this related question on our sister Stack Exchange "Can I get Speech Recognition in language other than English on Android?" Pocket Sphinx might be a possible route to go down. The answer there suggested that it only had Language Models and Dictionary Files for English languages - perhaps that has improved since the answer was given a couple of ...


1

It is quite hard, unless you have a lot of data. You'd better use radio alphabet "alpha", "bravo", "charlie".


1

Try to add the Cloud SDK distribution URI as a package source. There's more information about how to do that on 1. It should work just fine.


1

As mentioned, there is no pre-compiled binaries for the PI. However, installing from the tar.gz works just as well as from apt-get. Follow instructions here: "Versioned Archives" https://cloud.google.com/sdk/downloads


1

You are doing too much things with a command line. After you recognized you need to process the data, you need to handle errors and so on. It is much better to write your server with pocketsphinx API in C or in Python or Ruby, it will just be 20-30 lines of code.


1

No, most USB webcams have a built in audio card. The USB audio cards you see people referring to are one's like these: These are for if you want 3.5mm jack audio input. If you had a microphone with a 3.5mm jack. They also have a 3.5mm output.


1

I found the problem and fixed it. Using pyaudio's get_device_info_by_index and looking at the Microphone class in SpeechRecognition i was able to determine that it was defaulting to the wrong device number. Also using pyaudio's record.py sample i was also able to determine the 'CHUNK' size of 1024 was too much as well. So in my code i changed it to: ...


1

So, I tweaked my settings and for some reason the Google Speech API likes the Unsigned 8bit 11025Hz recording so I changed my code to be the following and now it works: #!/bin/bash echo "Recording..." arecord -D "plughw:1,0" -f U8 -r 11025 -d 5 | flac - -f --best --sample-rate 11025 -o file.flac echo "Processing..." wget -U "Mozilla/5.0" --post-file file....


Only top voted, non community-wiki answers of a minimum length are eligible