Streaming — ElevenLabs Documentation

The ElevenLabs API supports real-time audio streaming for select endpoints, returning raw audio bytes (e.g., MP3 data) directly over HTTP using chunked transfer encoding. This allows clients to process or play audio incrementally as it is generated.

Our official Node and Python libraries include utilities to simplify handling this continuous audio stream.

Streaming is supported for the Text to Speech API, Voice Changer API & Audio Isolation API. This section focuses on how streaming works for requests made to the Text to Speech API.

In Python, a streaming request looks like:

1 from elevenlabs import stream
2 from elevenlabs.client import ElevenLabs
3 
4 client = ElevenLabs()
5 
6 audio_stream = client.text_to_speech.convert_as_stream(
7     text="This is a test",
8     voice_id="JBFqnCBsd6RMkjVDRZzb",
9     model_id="eleven_multilingual_v2"
10 )
11 
12 # option 1: play the streamed audio locally
13 stream(audio_stream)
14 
15 # option 2: process the audio bytes manually
16 for chunk in audio_stream:
17     if isinstance(chunk, bytes):
18         print(chunk)

In Node / Typescript, a streaming request looks like:

1 import { ElevenLabsClient, stream } from 'elevenlabs';
2 import { Readable } from 'stream';
3 
4 const client = new ElevenLabsClient();
5 
6 async function main() {
7   const audioStream = await client.textToSpeech.convertAsStream('JBFqnCBsd6RMkjVDRZzb', {
8     text: 'This is a test',
9     model_id: 'eleven_multilingual_v2',
10   });
11 
12   // option 1: play the streamed audio locally
13   await stream(Readable.from(audioStream));
14 
15   // option 2: process the audio manually
16   for await (const chunk of audioStream) {
17     console.log(chunk);
18   }
19 }
20 
21 main();

1	from elevenlabs import stream
2	from elevenlabs.client import ElevenLabs
3
4	client = ElevenLabs()
5
6	audio_stream = client.text_to_speech.convert_as_stream(
7	text="This is a test",
8	voice_id="JBFqnCBsd6RMkjVDRZzb",
9	model_id="eleven_multilingual_v2"
10	)
11
12	# option 1: play the streamed audio locally
13	stream(audio_stream)
14
15	# option 2: process the audio bytes manually
16	for chunk in audio_stream:
17	if isinstance(chunk, bytes):
18	print(chunk)

1	import { ElevenLabsClient, stream } from 'elevenlabs';
2	import { Readable } from 'stream';
3
4	const client = new ElevenLabsClient();
5
6	async function main() {
7	const audioStream = await client.textToSpeech.convertAsStream('JBFqnCBsd6RMkjVDRZzb', {
8	text: 'This is a test',
9	model_id: 'eleven_multilingual_v2',
10	});
11
12	// option 1: play the streamed audio locally
13	await stream(Readable.from(audioStream));
14
15	// option 2: process the audio manually
16	for await (const chunk of audioStream) {
17	console.log(chunk);
18	}
19	}
20
21	main();