File size: 4,051 Bytes
9860c63 5956cea eed3dce ff39d68 9860c63 560a994 ff39d68 560a994 d250ad6 e839479 d250ad6 55a7c39 9860c63 abeede4 ff39d68 c6df2b8 ff39d68 560a994 d26fdac 3c577a6 6f488a4 560a994 b29c07d 7ea20fe 560a994 7ea20fe 560a994 c108596 bbc6b50 c108596 f9c9210 35b1732 f9c9210 0b49090 f9c9210 55a7c39 f9c9210 0b49090 55a7c39 0b49090 f9c9210 c108596 ab0b158 19a1b5d 0b49090 55a7c39 0b49090 560a994 358709b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 |
import os
os.system("pip install gradio==3.0.18")
from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification, AutoModelForTokenClassification
import gradio as gr
import spacy
nlp = spacy.load('en_core_web_sm')
##Speech Recognition
asr = pipeline("automatic-speech-recognition", "facebook/wav2vec2-base-960h")
def transcribe(audio):
text = asr(audio)["text"]
return text
def speech_to_text(speech):
text = asr(speech)["text"]
return text
##Summarization
summarizer = pipeline("summarization", model="knkarthick/MEETING_SUMMARY")
def summarize_text(text):
resp = summarizer(text)
stext = resp[0]['summary_text']
return stext
##Fiscal Tone Analysis
#fin_model = pipeline("text-classification", model="demo-org/auditor_review_model",
# tokenizer="demo-org/auditor_review_model",use_auth_token=auth_token)
fin_model= pipeline("sentiment-analysis", model='yiyanghkust/finbert-tone', tokenizer='yiyanghkust/finbert-tone')
def text_to_sentiment(text):
sentiment = fin_model(text)[0]["label"]
return sentiment
##Company Extraction
def fin_ner(text):
api = gr.Interface.load("dslim/bert-base-NER", src='models')
replaced_spans = api(text)
return replaced_spans
##Fiscal Sentiment by Sentence
def fin_ext(text):
doc = nlp(text)
doc_sents = [sent for sent in doc.sents]
sents_list = []
for sent in doc.sents:
sents_list.append(sent.text)
results = fin_model(sents_list)
results_list = []
for i in range(len(results)):
results_list.append(results[i]['label'])
fin_spans = []
fin_spans = list(zip(sents_list,results_list))
return fin_spans
##Forward Looking Statement
def fls(text):
doc = nlp(text)
doc_sents = [sent for sent in doc.sents]
sents_list = []
for sent in doc.sents:
sents_list.append(sent.text)
fls_model = pipeline("text-classification", model="yiyanghkust/finbert-fls", tokenizer="yiyanghkust/finbert-fls")
results = fls_model(sents_list)
results_list = []
for i in range(len(results)):
results_list.append(results[i]['label'])
fls_spans = []
fls_spans = list(zip(sents_list,results_list))
return fls_spans
demo = gr.Blocks()
with demo:
with gr.Row():
with gr.Column():
audio_file = gr.inputs.Audio(source="microphone", type="filepath")
with gr.Row():
b1 = gr.Button("Recognize Speech")
with gr.Row():
text = gr.Textbox(value="US retail sales fell in May for the first time in five months, restrained by a plunge in auto purchases and other big-ticket items, suggesting moderating demand for goods amid decades-high inflation. The value of overall retail purchases decreased 0.3%, after a downwardly revised 0.7% gain in April, Commerce Department figures showed Wednesday. Excluding vehicles, sales rose 0.5% last month. The figures aren’t adjusted for inflation.")
b1.click(speech_to_text, inputs=audio_file, outputs=text)
with gr.Row():
b2 = gr.Button("Summarize Text")
stext = gr.Textbox()
b2.click(summarize_text, inputs=text, outputs=stext)
with gr.Row():
b3 = gr.Button("Classify Financial Tone")
label = gr.Label()
b3.click(text_to_sentiment, inputs=stext, outputs=label)
with gr.Column():
b5 = gr.Button("Financial Tone and Forward Looking Statement Analysis")
with gr.Row():
fin_spans = gr.HighlightedText()
b5.click(fin_ext, inputs=text, outputs=fin_spans)
with gr.Row():
fls_spans = gr.HighlightedText()
b5.click(fls, inputs=text, outputs=fls_spans)
with gr.Row():
b4 = gr.Button("Identify Companies & Locations")
replaced_spans = gr.HighlightedText()
b4.click(fin_ner, inputs=text, outputs=replaced_spans)
demo.launch() |