ByteDance Seed Audio 1.0

ByteDance Seed Audio 1.0 makes audio from a text prompt, in a preset voice or guided by up to three reference clips or a picture.

Seed Audio 1.0 is ByteDance Seed's audio creation model. It makes audio from a text prompt, which can be guided by reference audio clips or a reference picture: the prompt is the text to say or a description of the audio to make, so it is not limited to reading a text aloud.

Write the text to say, or describe the audio, in prompt (up to 1,000 characters). voice picks a preset voice by name (the list is in the field help); audios takes up to three reference clips, each up to 30 seconds and 10 MB (WAV, MP3, PCM or Ogg Opus), that you point to in the prompt as @Audio1, @Audio2 and @Audio3; image is a reference picture and cannot be sent with audio clips. speed (0.5-2), pitch (-12 to 12 semitones) and volume (0.5-2) shape the delivery; multilingual suits non-English or mixed-language text; output_format is mp3 (the default), wav or ogg_opus.

Inputs: prompt (required), voice, audios, image, speed, pitch, volume, multilingual and output_format. Outputs: audio and duration (the length of the audio in seconds). Use a real person's voice or photo as a reference only with their permission.

Price: billed on what the run uses, at list: $0.1875 per minute of audio made, counted to the hundredth of a minute (a 5-second clip costs under two cents). You do not set the length: it follows the text and the speed. The price shown before the run is held on your balance and is worked out from the length of the prompt at a slow pace of 250 characters a minute (and the speed), which suits speech: $0.0375 at least and $1.50 at most. A description of a scene can make longer audio than its length suggests (the model refuses audio that is too long); you are charged exactly what the run cost. It makes one audio per request, does not make music and does not take a cloned-voice id.

Price: From $0.0375 per run, depending on prompt and speed. One run with default settings costs $0.0975.

Run ByteDance Seed Audio 1.0 via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/bytedance-seed-audio-1_0/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Hello from FlyMyAI."}'

API documentation · Create a free account · Pricing

Related models