Seeking detailed parameter docs on Wav2Vec via API

Beginners
1

Hi y’all I’m trying to find docs on how to call the Wav2Vec model for TTS via the API.

  1. The detailed API docs on parameters don’t seem yet to have information on the format of the API request for that model
  2. Trying to infer it from the demo page fails because of a CORS error in Chrome:
Access to fetch at 'https://api-audio-frontend.huggingface.co/models/facebook/wav2vec2-large-960h-lv60-self' from origin 'https://huggingface.co' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource. If an opaque response serves your needs, set the request's mode to 'no-cors' to fetch the resource with CORS disabled.

bundle.7995df3.js:1 POST https://api-audio-frontend.huggingface.co/models/facebook/wav2vec2-large-960h-lv60-self net::ERR_FAILED
run_api @ bundle.7995df3.js:1
handleClick @ bundle.7995df3.js:1
async function (async)
handleClick @ bundle.7995df3.js:1
(anonymous) @ bundle.7995df3.js:1
2

cc @patrickvonplaten

3

Wav2Vec2 is not yet included in the official inference-api - it should be included soon though :slight_smile:

4

Ah ok thanks @patrickvonplaten – sign me up for notifications when it is! Very interested in pricing too. The STT marketplace is rapidly maturing so I’m looking for the next price-driven challenger in this space.

1 Like
5

ASR was added to the list of parameters. You basically need to send the raw audio file.

Cheers,
Nicolas