Italian text-to-speech

Voices and samples

Every sample is this sentence, generated through the API with default settings: “Oggi c'è un tempo bellissimo, quindi nel pomeriggio andrò a passeggiare al parco con degli amici.”

Round-trip check: we transcribed each sample with Whisper Large v3 and compared it with the text we sent. It tests whether the speech is intelligible to a machine, not whether it sounds natural to you; for that, listen.

Quickstart

curl https://api.inferenceapis.com/v1/audio/speech \
  -H "Authorization: Bearer $INFERENCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "hexgrad/Kokoro-82M", "voice": "if_sara", "input": "Oggi c'\''è un tempo bellissimo, quindi nel pomeriggio andrò a passeggiare al parco con degli amici.", "response_format": "mp3"}' \
  --output italian.mp3

import os
from openai import OpenAI

client = OpenAI(base_url="https://api.inferenceapis.com/v1", api_key=os.environ["INFERENCE_API_KEY"])
audio = client.audio.speech.create(model="hexgrad/Kokoro-82M", voice="if_sara", input="Oggi c\'è un tempo bellissimo, quindi nel pomeriggio andrò a passeggiare al parco con degli amici.")
audio.write_to_file("italian.mp3")

Notes for Italian

• One female and one male voice.
• Numbers, dates and abbreviations are read as written. Spell out anything that must be pronounced a specific way.
• Pacing follows punctuation; there is no SSML. speed changes the overall rate.

Cost and limits

Kokoro is billed at $5.20 per million Unicode characters of input. The 97-character sample above is 5.4 seconds of audio and costs $0.0005; at that rate an hour of Italian speech is about 65,000 characters, or roughly $0.34. A request accepts up to 20,000 characters; for longer text, split on sentence boundaries and join the audio. A paragraph returns in under a second in our measurement. Orpheus is English-only, so Italian always uses Kokoro.

Other languages: English · Japanese · Chinese · Spanish · French · Hindi · Portuguese

Speech-to-text in Italian

The reverse direction: Whisper Large v3 transcribed our Italian test clips with 2.7% word error. See Italian speech-to-text.