Speech To Speech
Use Speech to Speech API to transform uploaded speech so it sounds like it was spoken by another voice. STS gives you full control over the emotions, timing and delivery.
API Features
High-quality voices
1000s of voices, in 29 languages, for every use-case, at 128kbps
Ultra-low latency
Achieve ~400ms audio generation times with our Turbo model.
Say it how you want it
Full control over emotions, tone, and pronunciation.
Audio generation
Generating speech-to-speech involves a similar process to text-to-speech, but with some adjustments in the API parameters. Instead of providing text when calling the API, you provide the path to an audio file that you would like to convert from one voice to another. Here’s a modified version of your code to illustrate how to generate speech-to-speech using the given API:
# Import necessary libraries
import requests # Used for making HTTP requests
import json # Used for working with JSON data
# Define constants for the script
CHUNK_SIZE = 1024 # Size of chunks to read/write at a time
XI_API_KEY = "<xi-api-key>" # Your API key for authentication
VOICE_ID = "<voice-id>" # ID of the voice model to use
AUDIO_FILE_PATH = "<path>" # Path to the input audio file
OUTPUT_PATH = "output.mp3" # Path to save the output audio file
# Construct the URL for the Speech-to-Speech API request
sts_url = f"https://api.elevenlabs.io/v1/speech-to-speech/{VOICE_ID}/stream"
# Set up headers for the API request, including the API key for authentication
headers = {
"Accept": "application/json",
"xi-api-key": XI_API_KEY
}
# Set up the data payload for the API request, including model ID and voice settings
# Note: voice settings are converted to a JSON string
data = {
"model_id": "eleven_english_sts_v2",
"voice_settings": json.dumps({
"stability": 0.5,
"similarity_boost": 0.8,
"style": 0.0,
"use_speaker_boost": True
})
}
# Set up the files to send with the request, including the input audio file
files = {
"audio": open(AUDIO_FILE_PATH, "rb")
}
# Make the POST request to the STS API with headers, data, and files, enabling streaming response
response = requests.post(sts_url, headers=headers, data=data, files=files, stream=True)
# Check if the request was successful
if response.ok:
# Open the output file in write-binary mode
with open(OUTPUT_PATH, "wb") as f:
# Read the response in chunks and write to the file
for chunk in response.iter_content(chunk_size=CHUNK_SIZE):
f.write(chunk)
# Inform the user of success
print("Audio stream saved successfully.")
else:
# Print the error message if the request was not successful
print(response.text)
Voices
We offer 1000s of voices in 29 languages. Visit the Voice Lab to explore our pre-made voices or clone your own. Visit the Voices Library to see voices generated by ElevenLabs users.
Supported languages
Our STS API is multilingual and currently supports the following languages:
Chinese, Korean, Dutch, Turkish, Swedish, Indonesian, Filipino, Japanese, Ukrainian, Greek, Czech, Finnish, Romanian, Russian, Danish, Bulgarian, Malay, Slovak, Croatian, Classic Arabic, Tamil, English, Polish, German, Spanish, French, Italian, Hindi and Portuguese.
To use them, simply provide the input audio in the language of your choice.
Headers
Your API key. This is required by most endpoints to access our API programatically. You can view your xi-api-key using the 'Profile' tab on the website.
Path Parameters
Voice ID to be used, you can use https://api.elevenlabs.io/v1/voices to list all the available voices.
Query Parameters
You can turn on latency optimizations at some cost of quality. The best possible final latency varies by model. Possible values: 0 - default mode (no latency optimizations) 1 - normal latency optimizations (about 50% of possible latency improvement of option 3) 2 - strong latency optimizations (about 75% of possible latency improvement of option 3) 3 - max latency optimizations 4 - max latency optimizations, but also with text normalizer turned off for even more latency savings (best latency, but can mispronounce eg numbers and dates).
Defaults to 0.
Output format of the generated audio. Must be one of: mp3_22050_32 - output format, mp3 with 22.05kHz sample rate at 32kbps. mp3_44100_32 - output format, mp3 with 44.1kHz sample rate at 32kbps. mp3_44100_64 - output format, mp3 with 44.1kHz sample rate at 64kbps. mp3_44100_96 - output format, mp3 with 44.1kHz sample rate at 96kbps. mp3_44100_128 - default output format, mp3 with 44.1kHz sample rate at 128kbps. mp3_44100_192 - output format, mp3 with 44.1kHz sample rate at 192kbps. Requires you to be subscribed to Creator tier or above. pcm_16000 - PCM format (S16LE) with 16kHz sample rate. pcm_22050 - PCM format (S16LE) with 22.05kHz sample rate. pcm_24000 - PCM format (S16LE) with 24kHz sample rate. pcm_44100 - PCM format (S16LE) with 44.1kHz sample rate. Requires you to be subscribed to Pro tier or above. ulaw_8000 - μ-law format (sometimes written mu-law, often approximated as u-law) with 8kHz sample rate. Note that this format is commonly used for Twilio audio inputs.
Body
The audio file which holds the content and emotion that will control the generated speech.
Identifier of the model that will be used, you can query them using GET /v1/models. The model needs to have support for speech to speech, you can check this using the can_do_voice_conversion property.
Voice settings overriding stored setttings for the given voice. They are applied only on the given request. Needs to be send as a JSON encoded string.
If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result. Determinism is not guaranteed.
Response
The response is of type file.