tts timestamps issue

[callMeMakerRen]

i see that in https://fal.ai/models/fal-ai/elevenlabs/tts/multilingual-v2 api doc, it says it support ,but i can not make it work , the api still return only mp3 data i think.

timestamps boolean
Whether to return timestamps for each word in the generated speech

[actuallyzefe]

Hi @callMeMakerRen,
Could you please provide a minimal reproduction example?

Also, can you confirm whether the following code works for you?

async function main() {
  const result = await fal.subscribe("fal-ai/elevenlabs/tts/multilingual-v2", {
    input: {
      text: "Hello! This is a test of the text to speech system, powered by ElevenLabs. How does it sound?",
      timestamps: true
    },
  });
  console.log(result.data.timestamps);

}

main();

Run this in your terminal:
export FAL_KEY=<your_fal_api_key> && tsx path-to-file

[callMeMakerRen]

the code you provide is work, but in the doc ,it says it support stream api ,so i try the below code ,and looks like all data is the mp3 data,

  const stream = await fal.stream("fal-ai/elevenlabs/tts/multilingual-v2", {
          input: {
            text: "Hello! This is a test of the text to speech system, powered by ElevenLabs. How does it sound?",
            timestamps: true
          },
      });
    
    stream.on("data", (data) => {
      console.log("data:", msg);
    });
    
    stream.on("done", () => {
      console.log("done");
    });  
      
    const result = await stream.done();
    console.log("result",result)

and i also wonder if i can use something like below as the doc says, not use the @fal-ai/client lib

curl --request POST \
  --no-buffer \
  --url https://fal.run/fal-ai/elevenlabs/tts/multilingual-v2/stream \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
     "text": "Hello! This is a test of the text to speech system, powered by ElevenLabs. How does it sound?",
     "timestamps": true
   }'