app.py

import openai, os
import gradio as gr
import time
import boto3
import json
import numpy as np
import wave
import io
import os
from langchain import OpenAI
from langchain.chains import ConversationChain
from langchain.memory import ConversationSummaryBufferMemory
from langchain.memory import ConversationBufferWindowMemory
from langchain.chat_models import ChatOpenAI
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain.schema import HumanMessage
import subprocess
from contextlib import closing
import asyncio
import requests

# This example uses aiofile for asynchronous file reads.
# It's not a dependency of the project but can be installed
# with `pip install aiofile`.
import aiofile

from amazon_transcribe.client import TranscribeStreamingClient
from amazon_transcribe.handlers import TranscriptResultStreamHandler
from amazon_transcribe.model import TranscriptEvent
from amazon_transcribe.utils import apply_realtime_delay


def run_shell_cmd(command):
    # Run a shell command
    result = subprocess.run(command, shell=True, capture_output=True, text=True)
    
    # Check the command output
    if result.returncode == 0:
        print("Command executed successfully")
        print("Command output:")
        print(result.stdout)
    else:
        print("Command failed")
        print("Error message:")
        print(result.stderr)

def wav_to_pcm(input_file, output_file):
    cmd = "ffmpeg -i " + input_file + " -f s16le -ar 16000 -ac 1 -acodec pcm_s16le " + output_file
    run_shell_cmd(cmd)
    

openai.api_key = os.environ["OPENAI_API_KEY"]
did_api_key = os.environ["DID_API_KEY"]
avatar_url = "https://create-images-results.d-id.com/DefaultPresenters/Magen_f/image.jpeg"

polly = boto3.client('polly', region_name='us-east-1')
s3 = boto3.client('s3')
transcribe = boto3.client('transcribe')


#memory = ConversationSummaryBufferMemory(llm=ChatOpenAI(), max_token_limit=2048)
memory = ConversationBufferWindowMemory(k=5)
conversation = ConversationChain(
    llm=OpenAI(streaming=True, callbacks=[StreamingStdOutCallbackHandler()], max_tokens=2048, temperature=0.5), 
    memory=memory,
)

SAMPLE_RATE = 16000
BYTES_PER_SAMPLE = 2
CHANNEL_NUMS = 1
AUDIO_PATH = ''
CHUNK_SIZE = 1024 * 8
REGION = "us-west-2"
transcript_text = ''
transcriptions = []

class MyEventHandler(TranscriptResultStreamHandler):
    def __init__(self, transcript_result_stream):
            super().__init__(transcript_result_stream)
            self.transcriptions = []
    async def handle_transcript_event(self, transcript_event: TranscriptEvent):
        # This handler can be implemented to handle transcriptions as needed.
        # Here's an example to get started.
        results = transcript_event.transcript.results
        for result in results:
            for alt in result.alternatives:
                print(alt.transcript)
                transcriptions.append(alt.transcript)


async def basic_transcribe():
    # Setup up our client with our chosen AWS region
    client = TranscribeStreamingClient(region=REGION)

    # Start transcription to generate our async stream
    stream = await client.start_stream_transcription(
        language_code="zh-CN",
        media_sample_rate_hz=SAMPLE_RATE,
        media_encoding="pcm",
    )

    async def write_chunks():
        # NOTE: For pre-recorded files longer than 5 minutes, the sent audio
        # chunks should be rate limited to match the realtime bitrate of the
        # audio stream to avoid signing issues.
        async with aiofile.AIOFile(AUDIO_PATH, "rb") as afp:
            reader = aiofile.Reader(afp, chunk_size=CHUNK_SIZE)
            await apply_realtime_delay(
                stream, reader, BYTES_PER_SAMPLE, SAMPLE_RATE, CHANNEL_NUMS
            )
        await stream.input_stream.end_stream()

    # Instantiate our handler and start processing events
    handler = MyEventHandler(stream.output_stream)
    await asyncio.gather(write_chunks(), handler.handle_events())
    # Retrieve the transcriptions from the handler
    #transcriptions = handler.transcriptions


def download_file(bucket_name, object_key, file_path):
    try:
        # Download the file from S3
        s3.download_file(bucket_name, object_key, file_path)
        print(f"File downloaded successfully: {file_path}")
    except Exception as e:
        print(f"Error downloading file: {str(e)}")

def polly_text_to_audio(audio_file_name, text, audio_format):

    if os.path.exists(audio_file_name):
        os.remove(audio_file_name)
        print("output mp3 file deleted successfully.")
    else:
        print("output mp3 file does not exist.")

    polly_response = polly.synthesize_speech(
        Text=text,
        OutputFormat=audio_format,  
        SampleRate='16000',
        VoiceId='Zhiyu',
        LanguageCode='cmn-CN',
        Engine='neural',
        LexiconNames=['tigoCN']
    )   
    
    # Access the audio stream from the response
    if "AudioStream" in polly_response:
        # Note: Closing the stream is important because the service throttles on the
        # number of parallel connections. Here we are using contextlib.closing to
        # ensure the close method of the stream object will be called automatically
        # at the end of the with statement's scope.
            with closing(polly_response["AudioStream"]) as stream:
               try:
                # Open a file for writing the output as a binary stream
                    with open(audio_file_name, "wb") as file:
                       file.write(stream.read())
               except IOError as error:
                  # Could not write to file, exit gracefully
                  print(error)
                  sys.exit(-1)

    else:
        # The response didn't contain audio data, exit gracefully
        print("Could not stream audio")
        sys.exit(-1)


def play_s3_voice(text):

    output_file = "/tmp/response.mp3"
    polly_text_to_audio(output_file, text, "mp3")

    # Upload the file to an S3 bucket
    audio_output_bucket_name = "lingo-audio-materials"
    audio_output_s3_key = "answers/response.mp3"

    s3.upload_file(output_file, audio_output_bucket_name, audio_output_s3_key)
    
    # Construct the S3 bucket URI
    s3_uri = f"s3://{audio_output_bucket_name}/{audio_output_s3_key}"

    print("audio output bucket name:"+audio_output_bucket_name)
    print("audio output key name:"+audio_output_s3_key)

    mp3_pre_signed_url = s3.generate_presigned_url('get_object',Params={'Bucket': audio_output_bucket_name,'Key': audio_output_s3_key},ExpiresIn=3600)
    print("mp3_pre_signed_url:"+mp3_pre_signed_url)

    current_dir = os.getcwd()
    print("current dir:"+current_dir)
    print("output_file_location: "+output_file)
    return output_file, mp3_pre_signed_url

def generate_talk_with_audio(input, avatar_url, api_key = did_api_key):
    url = "https://api.d-id.com/talks"
    payload = {
        "script": {
            "type": "audio",
            "audio_url": input
        },
        "config": {
            "auto_match": "true",
            "result_format": "mp4"
        },
        "source_url": avatar_url
    }
    headers = {
        "accept": "application/json",
        "content-type": "application/json",
        "authorization": "Basic " + api_key
    }

    response = requests.post(url, json=payload, headers=headers)
    return response.json()


def get_a_talk(id, api_key = os.environ.get('DID_API_KEY')):
    url = "https://api.d-id.com/talks/" + id
    headers = {
        "accept": "application/json",
        "authorization": "Basic "+api_key
    }
    response = requests.get(url, headers=headers)
    return response.json()

def get_mp4_video(input, avatar_url=avatar_url):
    response = generate_talk_with_audio(input=input, avatar_url=avatar_url)
    talk = get_a_talk(response['id'])
    video_url = ""
    index = 0
    while index < 30:
        index += 1
        if 'result_url' in talk:    
            video_url = talk['result_url']
            return video_url
        else:
            time.sleep(1)
            talk = get_a_talk(response['id'])
    return video_url


def predict(input, history=[]):
    if input is not None:
        history.append(input)
        response = conversation.predict(input=input)    
        audio_file, pre_signed_url = play_s3_voice(response)

        video_url = get_mp4_video(input=pre_signed_url, avatar_url=avatar_url)
        video_html = f"""<video width="320" height="240" controls autoplay><source src="{video_url}" type="video/mp4"></video>"""
        history.append(response)
        responses = [(u,b) for u,b in zip(history[::2], history[1::2])]
        return responses, audio_file, video_html, history
    else:
        video_html = f'<img src="{avatar_url}" width="320" height="240" alt="John Carmack">'
        responses = [(u,b) for u,b in zip(history[::2], history[1::2])]
        return responses, audio_file, video_html, history        

def transcribe_func_new(audio):
    audio_file = open(audio, "rb")
    wav_file = audio_file.name
    print("wav_file: "+wav_file)
    #transcript = openai.Audio.transcribe("whisper-1", audio_file)
    #return transcript['text']
    
    pcm_file = os.path.splitext(wav_file)[0] + ".pcm"

    wav_to_pcm(wav_file, pcm_file)
    if os.path.exists(pcm_file):
        print("pcm file exists")
    else:
        print("pcm file does not exist")
    AUDIO_PATH=pcm_file

    loop = asyncio.get_event_loop()
    loop.run_until_complete(basic_transcribe())
    loop.close()

    transcript_text = transcriptions[-1]
    print("final transcribe script: "+transcript_text)
    return transcript_text    

def transcribe_func_old(audio):
    audio_file = open(audio, "rb")
    file_name = audio_file.name
    #file_directory = os.path.dirname(audio_file.name)
    print("audio_file: "+file_name)
    #transcript = openai.Audio.transcribe("whisper-1", audio_file)
    #return transcript['text']
    
    # Set up the job parameters
    job_name = "lingo-demo"
    text_output_bucket = 'lingo-text-material' #this bucket is in us-west-1
    text_output_key = 'transcriptions/question.json'
    text_output_key = 'transcriptions/'+job_name+'.json'
    language_code = 'zh-CN'

    # Upload the file to an S3 bucket
    audio_input_bucket_name = "lingo-audio-material"
    audio_input_s3_key = "questions/tmp-question-from-huggingface.wav"
    
    s3.upload_file(file_name, audio_input_bucket_name, audio_input_s3_key)
    
    # Construct the S3 bucket URI
    s3_uri = f"s3://{audio_input_bucket_name}/{audio_input_s3_key}"

    response = transcribe.list_transcription_jobs()
    
    # Iterate through the jobs and print their names
    for job in response['TranscriptionJobSummaries']:
        print(job['TranscriptionJobName'])
        if job['TranscriptionJobName'] == job_name:
            response = transcribe.delete_transcription_job(TranscriptionJobName=job_name)
            print("delete transcribe job response:"+str(response))

    # Create the transcription job
    response = transcribe.start_transcription_job(
        TranscriptionJobName=job_name,
        Media={'MediaFileUri': s3_uri},
        MediaFormat='wav',
        LanguageCode=language_code,
        OutputBucketName=text_output_bucket,
        OutputKey=text_output_key
    )
    
    print("start transcribe job response:"+str(response))
    job_name = response["TranscriptionJob"]["TranscriptionJobName"]
    
    # Wait for the transcription job to complete
    while True:
        status = transcribe.get_transcription_job(TranscriptionJobName=job_name)['TranscriptionJob']['TranscriptionJobStatus']
        if status in ['COMPLETED', 'FAILED']:
            break
        print("Transcription job still in progress...")
        time.sleep(1)
    
    # Get the transcript
    #transcript = transcribe.get_transcription_job(TranscriptionJobName=job_name)
    transcript_uri = transcribe.get_transcription_job(TranscriptionJobName=job_name)['TranscriptionJob']['Transcript']['TranscriptFileUri']
    print("transcript uri: " + str(transcript_uri))
    
    transcript_file_content = s3.get_object(Bucket=text_output_bucket, Key=text_output_key)['Body'].read().decode('utf-8')
    print(transcript_file_content)
    json_data = json.loads(transcript_file_content)

    # Extract the transcript value
    transcript_text = json_data['results']['transcripts'][0]['transcript']
    return transcript_text    

def process_audio(audio, history=[]):
    text = transcribe_func_old(audio)
    return predict(text, history)

with gr.Blocks(css="#chatbot{height:350px} .overflow-y-auto{height:500px}") as demo:
    chatbot = gr.Chatbot(elem_id="chatbot")
    state = gr.State([])

    with gr.Row():
        txt = gr.Textbox(show_label=False, placeholder="Enter text and press enter").style(container=False)
        
    with gr.Row():
        audio_input = gr.Audio(source="microphone", type="filepath",  label="Audio Input")

    with gr.Row():
        audio_output = gr.Audio(type="filepath", label="Audio Output", elem_id="speaker", interactive=False)

    with gr.Row():
        video = gr.HTML(f'<img src="{avatar_url}" width="320" height="240" alt="John Carmack">', live=False)

    txt.submit(predict, [txt, state], [chatbot, audio_output, video, state])
    audio_input.change(process_audio, [audio_input, state], [chatbot, audio_output, video, state])

demo.launch(debug=True)