TTS with Bark🐶 and FastApi

The project is pulled from the official project Repository Bark 🐶.

It takes text from the user as input and converts it to a speech and responds with link to audio_file.wav. It is built in fastApi , it has two parameters. data and language. (Please refer net section for api references.)

🔗 Links

Installation

git clone the project the URL : FastApi-Bark

  git clone https://github.com/JeetDarekar001/FastApi-Bark.git

Create a new python envorienment and install the depedencies. Run the below commands.

python -m venv barkenv
.\barkenv\Scripts\activate
cd FastApi-Bark
pip install -r app/requirements.txt

This project required GPU to work faster, on the CPU envorienments it takes time.

After Installation, run the below command, form FastApi-Bark directory.

uvicorn app.app:app --reload

Open your brwoser and load the link http://127.0.0.1:8000, you should see the below output.

{"Message":"Yeahh!! Server is Online."}

Finally its done.💫

API Reference

Root

  GET /

Parameter	Type	Description
None	None	Points to the root of the serever. Display's message “Yeahh!! Server is Online."

Api Root

  GET /api/

Parameter	Type	Description
None	None	Root of api endpoint. Displays Message "Welcome to Text to Speech Conversion with Bark."

Test to speech Conversion

    POST /api/text_to_speech/

Parameter	Type	Description
data	str	text that should be converted to speech. Input should not be empty however it responses with error code 422 if input is null or less then 5 character.
language	str	Output langugae of the speech. Bark supports multiple languages, some are Hindi, English, Spanish, Italian, Chinese etc. Please Refer to the documnetation for more details. Do not provide additional languahe that is not present in Bark documnetation.

API responses

Api reposnse on successfully converting text to speech is as follows.

{
  "speech": {
    "path": "./app/Audio_files/Audio_2023-04-28_21-29-17.wav",
    "status_code": 200,
    "filename": "speech.mp3",
    "send_header_only": false,
    "media_type": "audio/mpeg",
    "background": null,
    "raw_headers": [
      [
        "content-type",
        "audio/mpeg"
      ],
      [
        "content-disposition",
        "attachment; filename=\"speech.mp3\""
      ]
    ],
    "_headers": {
      "content-type": "audio/mpeg",
      "content-disposition": "attachment; filename=\"speech.mp3\""
    },
    "stat_result": null
  },
  "data": "Welcome to London!!"
}