Ujaran ke teks

Sistem Multi-Modal dengan OpenAI API

James Chapman

Curriculum Manager, DataCamp

Berikutnya...

Tujuan kursus

Model audio OpenAI
Moderasi teks
Studi kasus: chatbot dukungan pelanggan

Gambar yang menampilkan model audio, moderasi teks, dan studi kasus

Rekap...

from openai import OpenAI


# Create the OpenAI client
client = OpenAI(api_key="<OPENAI_API_TOKEN>")


# Create a request to the Chat Completions endpoint
response = client.chat.completions.create(

    model="gpt-4o-mini",
    messages=[{"role": "user", 
               "content": "What is the OpenAI API?"}]

)

Tidak perlu kunci API—sudah disiapkan untuk Anda 🎉

Rekap...

# Extract the content from the response
print(response.choices[0].message.content)

The OpenAI API is a cloud-based service provided by OpenAI that allows developers
to integrate advanced AI models into their applications.

OpenAI API tidak hanya untuk teks 🚀

Model audio OpenAI

Kemampuan ujaran ke teks:

Transkripsikan audio
Terjemahkan audio non‑Inggris
Mendukung mp3, mp4, mpeg, mpga, m4a, wav, dan webm (batas 25 MB)

Kasus penggunaan:

Transkrip rapat
Takarir video

Ikon yang menampilkan rekaman audio dan blok teks.

Memproses panggilan pelanggan

Memuat file audio

Contoh: transkripsikan meeting_recording.mp3

audio_file = open("meeting_recording.mp3", "rb")

Jika file ada di direktori lain

audio_file = open("path/to/file/meeting_recording.mp3", "rb")

Membuat transkripsi

Endpoint audio

audio_file= open("meeting_recording.mp3", "rb")


response = client.audio.transcriptions.create(

    model="whisper-1",

    file=audio_file

)


print(response)

Transcription(text="Welcome everyone to the June product monthly. We'll get started in...)

¹ https://platform.openai.com/docs/guides/speech-to-text

Teks transkrip

print(response.text)

Welcome everyone to the June product monthly. We'll get started in just a minute.
Alright, let's get started. Today's agenda will start with a spotlight from Chris
on the new mobile user onboarding flow, then we'll review how we're tracking on
our quarterly targets, and finally, we'll finish with another spotlight from Katie
who will discuss the upcoming branding updates...

Transkripsi audio non‑Inggris

Ikon yang menampilkan rekaman audio dan blok teks.

Alur transkripsi:

open() file audio
Kirim permintaan transkripsi
Ekstrak teks

Membuat terjemahan

audio_file = open("non_english_audio.m4a", "rb")


response = client.audio.translations.create(

    model="whisper-1",

    file=audio_file

)


print(response.text)

The search volume for keywords like A I has increased rapidly since the launch of
Cha GTP.

Kinerja transkripsi

Kinerja bisa sangat bervariasi, tergantung pada:
- Kualitas audio
- Bahasa audio
- Pengetahuan model tentang topik

Berbagai bahasa di seluruh dunia.

Ayo berlatih!

Sistem Multi-Modal dengan OpenAI API