Alibaba Cloud Model Studio:Python SDK

Recognize speech from a microphone

import os
import signal  # for keyboard events handling (press "Ctrl+C" to terminate recording)
import sys

import dashscope
import pyaudio
from dashscope.audio.asr import *
# China (Beijing): Replace {WorkspaceId} with your actual workspace ID. The configuration varies by region.
dashscope.base_websocket_api_url = "wss://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1"

mic = None
stream = None

# Set recording parameters
sample_rate = 16000  # sampling rate (Hz)
channels = 1  # mono channel
dtype = 'int16'  # data type
format_pcm = 'pcm'  # the format of the audio data
block_size = 3200  # number of frames per buffer


def init_dashscope_api_key():
    """
        Set your DashScope API-key. More information:
        https://github.com/aliyun/alibabacloud-bailian-speech-demo/blob/master/PREREQUISITES.md
    """

    if 'DASHSCOPE_API_KEY' in os.environ:
        dashscope.api_key = os.environ[
            'DASHSCOPE_API_KEY']  # load API-key from environment variable DASHSCOPE_API_KEY
    else:
        dashscope.api_key = '<your-dashscope-api-key>'  # set API-key manually


# Real-time speech recognition callback
class Callback(RecognitionCallback):
    def on_open(self) -> None:
        global mic
        global stream
        print('RecognitionCallback open.')
        mic = pyaudio.PyAudio()
        stream = mic.open(format=pyaudio.paInt16,
                          channels=1,
                          rate=16000,
                          input=True)

    def on_close(self) -> None:
        global mic
        global stream
        print('RecognitionCallback close.')
        stream.stop_stream()
        stream.close()
        mic.terminate()
        stream = None
        mic = None

    def on_complete(self) -> None:
        print('RecognitionCallback completed.')  # recognition completed

    def on_error(self, message) -> None:
        print('RecognitionCallback task_id: ', message.request_id)
        print('RecognitionCallback error: ', message.message)
        # Stop and close the audio stream if it is running
        if 'stream' in globals() and stream.active:
            stream.stop()
            stream.close()
        # Forcefully exit the program
        sys.exit(1)

    def on_event(self, result: RecognitionResult) -> None:
        sentence = result.get_sentence()
        if 'text' in sentence:
            print('RecognitionCallback text: ', sentence['text'])
            if RecognitionResult.is_sentence_end(sentence):
                print(
                    'RecognitionCallback sentence end, request_id:%s, usage:%s'
                    % (result.get_request_id(), result.get_usage(sentence)))


def signal_handler(sig, frame):
    print('Ctrl+C pressed, stop recognition ...')
    # Stop recognition
    recognition.stop()
    print('Recognition stopped.')
    print(
        '[Metric] requestId: {}, first package delay ms: {}, last package delay ms: {}'
        .format(
            recognition.get_last_request_id(),
            recognition.get_first_package_delay(),
            recognition.get_last_package_delay(),
        ))
    # Forcefully exit the program
    sys.exit(0)


# main function
if __name__ == '__main__':
    init_dashscope_api_key()
    print('Initializing ...')

    # Create the recognition callback
    callback = Callback()

    # Call recognition service by async mode, you can customize the recognition parameters, like model, format,
    # sample_rate
    recognition = Recognition(
        model='paraformer-realtime-v2',
        format=format_pcm,
        # 'pcm', 'wav', 'opus', 'speex', 'aac', or 'amr'. You can check the supported formats in the document.
        sample_rate=sample_rate,
        # 8000 or 16000 is supported.
        semantic_punctuation_enabled=False,
        callback=callback)

    # Start recognition
    recognition.start()

    signal.signal(signal.SIGINT, signal_handler)
    print("Press 'Ctrl+C' to stop recording and recognition...")
    # Create a keyboard listener until "Ctrl+C" is pressed

    while True:
        if stream:
            data = stream.read(3200, exception_on_overflow=False)
            recognition.send_audio_frame(data)
        else:
            break

    recognition.stop()

Recognize a local audio file

The audio file used in the example is: asr_example.wav.

import os
import time
from dashscope.audio.asr import *
# China (Beijing): Replace {WorkspaceId} with your actual workspace ID. The configuration varies by region.
dashscope.base_websocket_api_url = "wss://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1"

# If you have not configured the API key in the environment variable, uncomment the following line of code and replace apiKey with your API key.
# import dashscope
# dashscope.api_key = "apiKey"

from datetime import datetime

def get_timestamp():
    now = datetime.now()
    formatted_timestamp = now.strftime("[%Y-%m-%d %H:%M:%S.%f]")
    return formatted_timestamp

class Callback(RecognitionCallback):
    def on_complete(self) -> None:
        print(get_timestamp() + ' Recognition completed')  # recognition complete

    def on_error(self, result: RecognitionResult) -> None:
        print('Recognition task_id: ', result.request_id)
        print('Recognition error: ', result.message)
        exit(0)

    def on_event(self, result: RecognitionResult) -> None:
        sentence = result.get_sentence()
        if 'text' in sentence:
            print(get_timestamp() + ' RecognitionCallback text: ', sentence['text'])
            if RecognitionResult.is_sentence_end(sentence):
                print(get_timestamp() + 
                    'RecognitionCallback sentence end, request_id:%s, usage:%s'
                    % (result.get_request_id(), result.get_usage(sentence)))


callback = Callback()

recognition = Recognition(model='paraformer-realtime-v2',
                          format='wav',
                          sample_rate=16000,
                          # The "language_hints" parameter is supported only by the paraformer-realtime-v2 model.
                          language_hints=['zh', 'en'],
                          callback=callback)

recognition.start()

try:
    audio_data: bytes = None
    f = open("asr_example.wav", 'rb')
    if os.path.getsize("asr_example.wav"):
        while True:
            audio_data = f.read(3200)
            if not audio_data:
                break
            else:
                recognition.send_audio_frame(audio_data)
            time.sleep(0.1)
    else:
        raise Exception(
            'The supplied file was empty (zero bytes long)')
    f.close()
except Exception as e:
    raise e

recognition.stop()

print(
    '[Metric] requestId: {}, first package delay ms: {}, last package delay ms: {}'
    .format(
        recognition.get_last_request_id(),
        recognition.get_first_package_delay(),
        recognition.get_last_package_delay(),
    ))