Quickstart

$
export SPEECHIFY_API_KEY="your-api-key-here"

$
pip install speechify-api

1
from speechify import Speechify
2
3
client = Speechify()  # uses SPEECHIFY_API_KEY env var
4
5
response = client.tts.audio.speech(
6
    input="Hello! This is my first Speechify API call.",
7
    voice_id="george",
8
    audio_format="mp3",
9
)
10
11
with open("output.mp3", "wb") as f:
12
    f.write(response.audio_data)
13
14
print(f"Audio saved ({len(response.audio_data)} bytes)")

$
# macOS
$
afplay output.mp3
$
$
# Linux
$
aplay output.mp3

Choose a voice

The API includes built-in voices. List them to find one that fits your use case:

1
voices = client.tts.voices.list()
2
3
for voice in voices:
4
    print(f"{voice.id}: {voice.display_name} ({voice.gender})")

Popular built-in voices: george, henry, carly, sophia.

You can also clone any voice from a short audio sample.

Add emotion

Use SSML to control how the voice sounds:

1
<speak>
2
  <speechify:style emotion="cheerful">
3
    Great news! Your order has been shipped!
4
  </speechify:style>
5
</speak>

Pass SSML as the input parameter — the API detects it automatically. See Emotion Control for the full list of 13 supported emotions.

Next steps