Python Create Webex Audio Prompt

If you are unable to use text to speech, here is a process using python to create an audio file compatible with WebEx Contact Center

The script creates an audio file with the exact telephony specifications required by WebEx Contact Center:

• Format: .wav (Waveform Audio File Format)
• Sample Rate: 8000 Hz (8 kHz)
• Channels: 1 (Mono)
• Audio Codec / Encoding: pcm_mulaw (G.711 μ-law, 8-bit) [1, 2, 3, 4]

In your terminal

• pip install edge-tts ffmpeg-python
• winget install -e –id Gyan.FFmpeg

Important – After install, close your terminal, CD back to the directory and run python script.py

import asyncio
import edge_tts
import ffmpeg

# 1. Define your configuration
TEXT_PROMPT = "Thank you for calling our support center. Please hold while we connect you."
OUTPUT_FILENAME = "webex_prompt.wav"
VOICE = "en-US-AvaNeural"  # High-quality natural English voice

async def text_to_webex_audio():
    temp_mp3 = "temp_voice.mp3"
    
    print("Generating neural text-to-speech...")
    communicate = edge_tts.Communicate(TEXT_PROMPT, VOICE)
    await communicate.save(temp_mp3)
    
    print("Converting to Webex specifications (8kHz, Mono, 8-bit u-law)...")
    try:
        (
            ffmpeg
            .input(temp_mp3)
            .output(
                OUTPUT_FILENAME,
                ar='8000',       # Sample Rate: 8kHz
                ac=1,            # Channels: Mono
                acodec='pcm_mulaw' # Encoding: u-law / G.711
            )
            .overwrite_output()
            .run(capture_stdout=True, capture_stderr=True)
        )
        print(f"Success! '{OUTPUT_FILENAME}' is ready to upload.")
    except ffmpeg.Error as e:
        print("FFmpeg Error:", e.stderr.decode())

# Run the process
asyncio.run(text_to_webex_audio())