richtier / alexa-voice-service-client Goto Github PK

Hello,

I want to use this library to generate some answer waves of a question set.
But I encounter http.client.NO_CONTENT.
I have successful generated the refresh token and register a product.
Could you give me some help on this issue? Thanks.

Traceback (most recent call last):
  File "test.py", line 8, in <module>
    alexa_client.connect()  # authenticate and other handshaking steps
  File "/home/mstts/Documents/Amazon/alexa-voice-service-client/avs_client/avs_client/client.py", line 43, in connect
    self.synchronise_device_state()
  File "/home/mstts/Documents/Amazon/alexa-voice-service-client/avs_client/avs_client/client.py", line 55, in synchronise_device_state
    device_state=self.device_manager.get_device_state(),
  File "/home/mstts/Documents/Amazon/alexa-voice-service-client/avs_client/avs_client/connection.py", line 72, in synchronise_device_state
    assert response.status == http.client.NO_CONTENT
AssertionError

I have a custom skill in alexa account. I have triggered it from microphone by following the instructions and executed --> alexa_client.demo.streaming_microphone along with required client parameters. I'm able to invoke the skill & getting the response for the first command, but from the second command it's triggering fallback. I tried to print the directive response to check the dialogRequestId, it is same.

I tried for "Alexa, Open twenty questions" skill too. Unable to play continuously. Connected is termination after responding to first question

https://pypi.org/project/alexa-client/#files doesnt contain an sdist, and there are no tags here, so there is no way to get a versioned tarball of this software. The wheel is unsuitable for distro packaging

Hello.
For example, I sent request as "countdown 10 seconds" and timer was created.
I got response as "10 seconds, starting now".
I checked that timer was created in Alexa.amazon.com.
So, I sent request and get response synchronously.
how to get other directives (asynch alert responses) ?
I don't get this directive.

I am pretty sure I did all steps right
still i am getting this error
Traceback (most recent call last):
File "alexa.py", line 8, in
client.connect() # authenticate and other handshaking steps
File "C:\Users\Harshit.dell-PC\AppData\Local\Programs\Python\Python36\lib\site-packages\alexa_client\alexa_client\client.py", line 33, in connect
self.synchronise_device_state()
File "C:\Users\Harshit.dell-PC\AppData\Local\Programs\Python\Python36\lib\site-packages\alexa_client\alexa_client\client.py", line 49, in synchronise_device_state
device_state=self.device_manager.get_device_state(),
File "C:\Users\Harshit.dell-PC\AppData\Local\Programs\Python\Python36\lib\site-packages\alexa_client\alexa_client\connection.py", line 74, in synchronise_device_state
assert response.status in [http.client.NO_CONTENT, http.client.OK]
AssertionError

I tried your example of using audio files but I get the error:
TypeError: 'NoneType' object is not iterable

when I run the script.
Is this bug known? And if so, how can I fix it?

How to get response as a text from Alexa ?

When I run
python ./avs_client/refreshtoken/serve.py \ --device-type-id=enter-device-type-id-here \ --client-id=enter-client-id-here \ --client-secret=enter-client-secret-here

I assume the --device-type-id is the "product id"
I have added "https://localhost:9000" to allowed origin and added "https://localhost:9000/callback/" to allowed return urls.

But when I run the server.py and go to http:localhost:9000.

It always give me the error.

We're sorry!
An error occurred when we tried to process your request. Rest assured, we're already working on the problem and expect to resolve it shortly.

Hide Details
Error Summary
400 Bad Request
The redirect URI you provided has not been whitelisted for your application. Please add your redirect URI in the 'Allowed Return URLs' section under 'Web Settings' for your Security Profile on Amazon Developer Portal.
Request Details
client_id=amzn1.application-oa2-client.31d6fe43e21d4554bceea25fd258e1d6
scope=alexa%3Aall
scope_data=%7B%22alexa%3Aall%22%3A+%7B%22productID%22%3A+%22bottled_joy%22%2C+%22productInstanceAttributes%22%3A+%7B%22deviceSerialNumber%22%3A+%22001%22%7D%7D%7D
response_type=code
redirect_uri=http%3A%2F%2F1.0.0.127.in-addr.arpa%3A9000%2Fcallback%2F

https://developer.amazon.com/zh/docs/alexa-voice-service/system.html
The response.status is always 400,
I can't change it to en-GB,
help me, please

Hi,

As per the following example of yours, the response is stored as an mp3 file. I want text as a response. How can I get that ?

with open('./tests/resources/alexa_what_time_is_it.wav', 'rb') as f:
resp = client.send_audio_file(f)
for i, directive in enumerate(resp):
if directive.name in ['Speak', 'Play']:
with open(f'./output_{i}.mp3', 'wb') as f:
f.write(directive.audio_attachment)

Hi, I'm seeing same error as #26. The python version is 2.7.9.

I'm able to get responses for generic statements as -

What is the time now
How is the weather

But I'm not able to use my custom skill. Any idea on how i can proceed?

Thanks for the amazing library though :)

Could not find a version that satisfies the requirement avs_client (from versions: )
No matching distribution found for avs_client

That's the error I keep getting whenever I run the install command

Hi Team,

I tried with Multi-step requests, but I'm getting 'NoneType' object has no attribute 'request' error. Below is my code.

Code

from alexa_client import AlexaClient
from alexa_client.alexa_client import helpers

client = AlexaClient(
client_id='amzn1.application-oa2-client.ce399dfcf5c958b66c5d0c4cd1b03',
secret='53a7c2f016764402f6c4bcf53ce7e8a0f47b9a37dfa46bf2fb31e3ccf099',
refresh_token='Atzr|IwEBIEiWGBIY5fgZPrjzbdLgJfGdHInC6gDOTgTTGuCMM2H9aQWQ-7MZFn9NUmZXn-B_IHrRhG1WAi4geiE6PbohWIY81cMtK0tFAkYzkhGMDQNNXDoVMsVk1rTNr354ZopD3IQ5r1WSko8XzHLir-RskaaUERKyqEsOIEvqYlozmP6VjELnAJvglQYawtJFSf24QUsF24jYROqUXdZ87eIudvWWMh-i37D25fWLQo230LzLf5D9IABshQs7nra3PFF2cacKUcRkZ3IL3op7738tPzxf4q32GGECsvi5lXQHk0NlYU7JgjIa-adoKSIToE9ZHV2kf2-XB5jdqlpelMBeoB6d8wH9XjXB4ZpP-RNNx9EEAaP9p7YJ04EMY',
)

dialog_request_id = helpers.generate_unique_id()
directives_one = client.send_audio_file('C:/Users/nitin.verma/PycharmProjects/AwsPolly/resources/alexa_what_time_is_it.wav', dialog_request_id=dialog_request_id)
directives_two = client.send_audio_file('C:/Users/nitin.verma/PycharmProjects/AwsPolly/resources/alexa_go_to_dark_forest.wav', dialog_request_id=dialog_request_id)
directives_three = client.send_audio_file('C:/Users/nitin.verma/PycharmProjects/AwsPolly/resources/alexa_harumph.wav', dialog_request_id=dialog_request_id)

Environment :
Python 3.6.9 on Ubuntu 16.04

I think this bug related to multipart from request. For each request I get the header name, but the payload is missing. When I print the whole directive response, it shows the payload but the Directive class can't get it

{'header': {'dialogRequestId': '....', 'messageId': .... 'name': 'RenderTemplate', 'namespace': 'TemplateRuntime'}, 'payload': {'skillIcon': ...... 'textField': 'this', 'title': {'mainTitle': 'Transcription', 'subTitle': 'Transcribe Test'}, 'token':....', 'type': 'BodyTemplate1'}}

Seems the Alexa return the payload, but the directive class can't get it

How to send text to Alexa instead of audio file? Is there any way?

Hello.
I am using the library and my compliments for this, is very well done.

By the way i have a problem:
The Post stream with audio events are working pretty good.
But after i am enabling the downchannel i am not anymore able to receive event on this, is that normal?
Did anyone find any solution about this?
If i try a read on the downchannel all the program is blocked

Thanks for caring,

Alessio

Currently, host is hardcoded to old legacy url 'avs-alexa-eu.amazon.com'. AVS enforcing to use specific region Base URLs https://developer.amazon.com/docs/alexa-voice-service/api-overview.html#endpoints. For example, my client expected be connected with host North America Host Base URLs.

if the Host url is input parameter along with client id, secret and refresh_token, then it will useful to your package as it is instead of forking a copy.

Hello, I got the authentication code but when i use the authentication code to fetch the access token i get 400 Client Error: Bad Request for url: https://api.amazon.com/auth/o2/token. DId i miss anything ?.

Originally posted by @s1782662 in #21 (comment)

When running the streaming_microphone demo (or the Alexa Browser client), the sync device state will error out with a failed assertion. Debug inspection showed it is returning an OK response.

→ python3  -m alexa_client.demo.streaming_microphone  --client-id="amzn1.application-**" --client-secret="**" --refresh-token='**'
listening. Press CTRL + C to exit.
Say something to Alexa.
Traceback (most recent call last):
  File "/anaconda3/lib/python3.7/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/anaconda3/lib/python3.7/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/alexa-voice-service-client/alexa_client/demo/streaming_microphone.py", line 84, in <module>
    refresh_token=parsed.refresh_token,
  File "/alexa-voice-service-client/alexa_client/demo/streaming_microphone.py", line 42, in main
    alexa_client.connect()
  File "/alexa-voice-service-client/alexa_client/alexa_client/client.py", line 35, in connect
    self.synchronise_device_state()
  File "/alexa-voice-service-client/alexa_client/alexa_client/client.py", line 51, in synchronise_device_state
    device_state=self.device_manager.get_device_state(),
  File "/alexa-voice-service-client/alexa_client/alexa_client/connection.py", line 77, in synchronise_device_state
    assert response.status in ( http.client.NO_CONTENT, )
AssertionError

I am using Azure text to speech to test this client. I used Azure to generate the voice of the following command with "riff-16khz-16bit-mono-pcm" output format.

"Alexa, play True or False",
"Alexa, start Song Quiz",
"Alexa, ask Baby Groot what time is it",
"Alexa, Open Eighties Letters",
"Alexa, ask Question of the Day.",
"Alexa, Play Jeopardy!",

"Alexa, play Twenty Questions.",
"Alexa, open Heads Up! and play Superstars",
"Alexa, play Geo Quiz",
"Alexa, Open Alien Simulator"

The first 6 commands return legit response.
I don't know why, but the last 4 commands always return an empty response.

This may not be related to this repo, but do you happen to know why?
I tried to change the voice volume and make Azure to speak it slower. None of this worked.

I'm attempting to create an AWS lambda function that calls the .connect() method and then asks a question. When I submit a single question at a time, it appears to work. When submitting multiple requests, I'm getting a "Stream forcefully closed" error in hyper.http20 when getting the response. Is it possible for the client id, client secret and refresh token to be used by multiple connections concurrently? Any thoughts on what might cause this error?

When running the sample code given in the readme, i get the following error:
Traceback (most recent call last): File "/root/Desktop/python/Project-Mark/trial.py", line 12, in <module> for i, directive in enumerate(client.send_audio_file(f)): TypeError: 'NoneType' object is not iterable

Hi Team,
I have used amazon polly to generate the wav file and passed it to the alexa client, but it's giving me the error as

for i, directive in enumerate(client.send_audio_file(f)):
TypeError: 'NoneType' object is not iterable

Amazon polly code:
import boto3
import os

os.environ['AWS_PROFILE'] = "NitinVermaAskCli"
os.environ['AWS_DEFAULT_REGION'] = "us-east-1"
polly_client = boto3.Session(aws_access_key_id='AKIAZYOFB32NZRDUP',
aws_secret_access_key='jTuAUFmtQte4LfG0QIVuHS+E0kh1qJ3XVBmw',
region_name='us-west-1').client('polly')

response = polly_client.synthesize_speech(VoiceId='Joanna',
OutputFormat='mp3',
Text = 'Alexa, ask doctor connect')

file = open('C:/Users/nitin.verma/PycharmProjects/AwsPolly/resources/alexa_doctor_connect.wav', 'wb')
file.write(response['AudioStream'].read())
file.close()

Alexa_Client
from alexa_client import AlexaClient

client = AlexaClient(
client_id='amzn1.application-oa2-client.ce399dfcf5cf8b66c5d0c4cd1b03',
secret='53a7c2f016764402f6c4bcf53ce7e8a0f4b9a37dfa46bf2fb31e3ccf099',
refresh_token='Atzr|IwEBIEiWGBIY5fgZPrjzbdLgJfGdLGYHHInC6gDOTgTTGuCMM2H9aQWQ-7MZFn9NUmZXn-B_IHrRhG1WAi4geiE6PbohWIY81cMtK0tFAkYzkhGMDQNNXDoVMsVk1rTNr354ZopD3IQ5r1WSko8XzHLir-RskaaUERKyqEsOIEvqYlozmP6VjELnAJvglQYawtJFSf24QvmoUsFqUXdZ87eIudvWWMh-i37D25fWLQo230LzLf5D9IABshQs7nra3PFF2cacKU1Q3IL3op7738tPzxf4q32GGECsvi5lXQHk0NlYU7JgjIa-adoKSIToE9ZHV2kf2-XB5jdqlpelMBeoB6d8wH9XjXB4ZpP-RNNx9EEAaP9p7YJ04EMY',
)
client.connect() # authenticate and other handshaking steps
with open('C:/Users/nitin.verma/PycharmProjects/AwsPolly/resources/alexa_doctor_connect.wav', 'rb') as f:
for i, directive in enumerate(client.send_audio_file(f)):
if directive.name in ['Speak', 'Play']:
with open(f'./output_{i}.mp3', 'wb') as f:
f.write(directive.audio_attachment)
print('done')

Requesting you to please help me out to resolve this.

I need to customized the profile configurations of connection, set it to NEAR_FIELD or FAR_FIELD but can not do it on current version.
refer https://developer.amazon.com/docs/alexa-voice-service/speechrecognizer.html#profile

i fork a branch is not a better choice.

suggest to add a class member named as profile, default set to 'CLOSE_TALK', so that we can set it
https://github.com/richtier/alexa-voice-service-client/blob/master/alexa_client/alexa_client/connection.py Line 99
'payload': {
'profile': 'CLOSE_TALK',
'format': 'AUDIO_L16_RATE_16000_CHANNELS_1'
}
Line 12
From
host = 'avs-alexa-eu.amazon.com'
To
host = 'avs-alexa-eu.amazon.com'
profile = 'CLOSE_TALK'

Currently the project does not support TemplateRuntime interface.

For devices where users want to display text and images the supports are needed.

The authentication API you've provided allows you to set the client_id, secret, and refresh_token for the product credentials, but there doesn't seem to be a way to bind my Amazon user credentials to this device (as you can with an e.g., an Echo Dot).

I'm attempting to use your Python module to test not-yet-published Alexa skills. Even though I've registered the built-in app to create this Python script using the same Amazon developer account as the skill, the app doesn't see it when I query. I assume the issue is that I haven't actually logged into this new "device" using my Amazon credentials.

It's fairly easy to get an access_token and refresh_token after authenticating with Amazon via code-based linking authorization (spawning a webpage, giving it a short code, etc.). I just need some way to hook this into this API.

Can you confirm whether there's a way to do what I want?

Currently during the multi-turn requests, the AVS client is not sending initiator.

But for supporting multi-turn AVS expects initiator (when it's being send with previous AVS response).

From AVS documentation

If present, the initiator object included in the payload of the ExpectSpeech directive must be passed back to Alexa as the initiator object in the following Recognize event. If initiator is absent from the payload, the following Recognize event should not include initiator.

Details can be found here https://developer.amazon.com/docs/alexa-voice-service/speechrecognizer.html#expectspeech

Could you make an option to change the language that Alexa is waiting for? I need to change it to Portuguese instead of English, How can I?

i installed and ran your audio file demo ... the program does not return however it does create the output mp3 file ... is this the correct or is there something wrong with my installation ?

Any idea on what can be done to get the transcript?