​
Install Dependencies

pip install pyaudio 
pip install reverie-sdk
pip install python-dotenv

​
How does it Works?

The script follows these main steps:

  1. Environment Setup:
    The script loads the required credentials (REVERIE_APP_ID and REVERIE_API_KEY) from environment variables using the dotenv library. These credentials are essential for authenticating API requests to the Reverie SDK.
Python
import os
import dotenv
from reverie_sdk import ReverieClient

dotenv.load_dotenv("../.env")
REVERIE_APP_ID = os.environ.get("REVERIE_APP_ID")
REVERIE_API_KEY = os.environ.get("REVERIE_API_KEY")
client = ReverieClient(
    api_key=REVERIE_API_KEY,
    app_id=REVERIE_APP_ID,
)
  1. Real-Time Audio Streaming:
    The script uses pyaudio to capture audio from the microphone in real-time. The captured audio is sent to the Reverie ASR service for continuous transcription via the AudioStream object. The audio data is streamed asynchronously, providing live transcription results.
Python
import pyaudio
import asyncio
from reverie_sdk.services.asr import AudioStream

stream = AudioStream()
pa = pyaudio.PyAudio()

def mic_callback(in_data, frame_count, time_info, status):
    try:
        asyncio.run(stream.add_chunk_async(in_data))
    except Exception as e:
        print(f"Error in mic callback: {e}")
        return (None, pyaudio.paAbort)
    return (None, pyaudio.paContinue)

pa_stream = pa.open(
    rate=16000,
    channels=1,
    format=pyaudio.paInt16,
    frames_per_buffer=1024,
    input=True,
    stream_callback=mic_callback,
)
pa_stream.start_stream()
  1. Speech-to-Text (STT) Conversion:
    The captured audio is processed by the Reverie ASR service, which transcribes the spoken words into text. The transcription happens in real-time, and each segment of the transcribed text is displayed as it is received.
Python
await client.asr.stt_stream_async(
    src_lang=source_lang,
    bytes_or_stream=stream,
    callback=asr_callback,
    format="16k_int16",
    punctuate="true",
)
  1. Callback Function for ASR Results:
    A callback function is used to handle the ASR responses. Once the transcription is finalized, the transcribed text is returned and printed.
Python
from reverie_sdk.services.asr import ReverieAsrResult

asr_responses: list[ReverieAsrResult] = []

def asr_callback(resp: ReverieAsrResult):
    print(resp)
    asr_responses.append(resp)

while True:
    if len(asr_responses) == 0:
        continue

    if asr_responses[-1].final:
        print(asr_responses[-1].display_text)
        return asr_responses[-1].display_text
  1. Text Translation:
    After obtaining the transcribed text, the script sends it to the Reverie NMT service for translation into the specified target language (e.g., from Hindi to English in this case). The translation result is then printed.
Python
src_lang = "hi"
tgt_lang = "en"

resp: str = asyncio.run(speech_to_text(src_lang))

if len(resp.strip()) > 0:
    tgt_resp = (
        client.nmt.localization(
            [resp],
            src_lang=src_lang,
            tgt_lang=tgt_lang,
            enableTransliteration=False,
            enableNmt=True,
        )
        .responseList[0]
        .outString
    )
    print(tgt_resp)
  1. Error Handling:
    Comprehensive error handling is included to ensure smooth operation, including managing microphone input errors and any potential issues during the transcription or translation processes.
Python
import traceback

try:
    # [Audio streaming and transcription code]
except Exception as err:
    print(f"Error in Speech-to-Text: {err}")
    traceback.print_exc()
finally:
    pa_stream.stop_stream()
    pa_stream.close()

​
Sample Code

import asyncio
import os
import traceback
import pyaudio
from reverie_sdk import ReverieClient
from reverie_sdk.services.asr import ReverieAsrResult, AudioStream
import dotenv

dotenv.load_dotenv("../.env")

REVERIE_APP_ID = os.environ.get("REVERIE_APP_ID")
REVERIE_API_KEY = os.environ.get("REVERIE_API_KEY")

client = ReverieClient(
    api_key=REVERIE_API_KEY,
    app_id=REVERIE_APP_ID,
)


async def speech_to_text(source_lang):
    stream = AudioStream()
    pa = pyaudio.PyAudio()

    def mic_callback(in_data, frame_count, time_info, status):
        try:
            asyncio.run(stream.add_chunk_async(in_data))
        except Exception as e:
            print(f"Error in mic callback: {e}")
            return (None, pyaudio.paAbort)
        return (None, pyaudio.paContinue)

    pa_stream = pa.open(
        rate=16000,
        channels=1,
        format=pyaudio.paInt16,
        frames_per_buffer=1024,
        input=True,
        stream_callback=mic_callback,
    )
    pa_stream.start_stream()

    print("Listening at STT..")

    asr_responses: list[ReverieAsrResult] = []

    def asr_callback(resp: ReverieAsrResult):
        print(resp)
        asr_responses.append(resp)

    try:
        await client.asr.stt_stream_async(
            src_lang=source_lang,
            bytes_or_stream=stream,
            callback=asr_callback,
            format="16k_int16",
            punctuate="true",
        )

        while True:
            if len(asr_responses) == 0:
                continue

            if asr_responses[-1].final:
                print(asr_responses[-1].display_text)
                return asr_responses[-1].display_text

    except Exception as err:
        print(f"Error in Speech-to-Text: {err}")
        traceback.print_exc()

    finally:
        pa_stream.stop_stream()
        pa_stream.close()


src_lang = "hi"
tgt_lang = "en"


resp: str = asyncio.run(speech_to_text(src_lang))

if len(resp.strip()) > 0:
    tgt_resp = (
        client.nmt.localization(
            [resp],
            src_lang=src_lang,
            tgt_lang=tgt_lang,
            enableTransliteration=False,
            enableNmt=True,
        )
        .responseList[0]
        .outString
    )
    print(tgt_resp)