FAL.ai Wizper

FAL.ai Wizper is a hosted, real-time variant of OpenAI’s Whisper v3 that powers Speech-to-Text and highly-accurate, on-the-fly translation.

With the Stream Python AI SDK you can use Wizper inside your video calls in just a few lines of code.

Installation

Install the Stream FAL.ai plugin with

uv add getstream-plugins-fal

Example

Check out our auto-translation example to see a practical implementation of the plugin and get inspiration for your own projects, or read on for some key details.

from getstream.plugins.fal import FalWizperSTT

# 1. Pure transcription (default)
stt = FalWizperSTT()

# 2. Translate everything participants say to Spanish
stt = FalWizperSTT(target_language="es")

@stt.on("transcript")
async def on_transcript(text: str, user: dict, metadata: dict):
    print(f"{user['name']} said → {text}")

# Send Stream PCM audio frames to the plugin
await stt.process_audio(pcm_data)

# Close when finished
await stt.close()

Initialisation

First, make sure you’ve created an API key for the FAL.ai service and set the FAL_KEY environment variable to your API key.

The FAL.ai plugin is exposed via the FalWizperSTT class:

from getstream.plugins.fal import FalWizperSTT

# 1. Pure transcription (default)
stt = FalWizperSTT()

# 2. Translate everything participants say to Spanish
stt = FalWizperSTT(target_language="es")

Parameters

You can customise the behaviour of Wizper through the following parameters:

Name	Type	Default	Description
`task`	`str\|None`	`transcribe`	Task to perform on the audio. Either transcribe or translate. Default value: “transcribe”.
`target_language`	`str\|None`	`None`	ISO-639-1 code. If set, Wizper translates the recognised text to this language.
`sample_rate`	`int`	`48000`	Incoming PCM sample rate in Hertz.

Functionality

Process Audio

After joining a Stream call, use a VAD (e.g. Silero) to capture speech events and pass them to Wizper for real-time STT (and optional translation):

from getstream.video import rtc

async with rtc.join(call, bot_user_id) as connection:

    @connection.on("audio")
    async def _on_audio(pcm: PcmData, user):
        await stt.process_audio(pcm, user)

Events

The plugin emits the standard Stream STT events:

Transcript Event

@stt.on("transcript")
async def _on_transcript(text: str, user: any, metadata: dict):
    ...

Error Event

@stt.on("error")
async def _on_error(error: Exception):
    # Handle errors returned by Wizper here

Close

When you’re done, close the Wizper connection with close():

await stt.close()

Silero

Create Your Own Plugin