Transcribing recorded audio in FlEt web browser.

[leony777]

Question

I'm having a hard time getting and transcribing audio when running the code with view=web_browser. It gives me a URL, but I want to get the file to transcribe it. I managed to do it on Windows, but I'm struggling with the web version.

Code sample

import flet as ft

async def main(page: ft.Page):
    page.horizontal_alignment = ft.CrossAxisAlignment.CENTER
    page.appbar = ft.AppBar(title=ft.Text("Audio Recorder"), center_title=True)

    path = "test-audio-file.wav"

    async def handle_start_recording(e):
        print(f"StartRecording: {path}")
        await audio_rec.start_recording_async(path)

    async def handle_stop_recording(e):

        ## ###################### HERE ####################################
        output_path = await audio_rec.stop_recording_async()
        print(f"StopRecording: {output_path}")

    async def handle_state_change(e):
        print(f"State Changed: {e.data}")

    audio_rec = ft.AudioRecorder(
        audio_encoder=ft.AudioEncoder.WAV,
        on_state_changed=handle_state_change,
    )
    page.overlay.append(audio_rec)
    await page.update_async()

    await page.add_async(
        ft.ElevatedButton("Start Audio Recorder", on_click=handle_start_recording),
        ft.ElevatedButton("Stop Audio Recorder", on_click=handle_stop_recording)
    )


ft.app(target=main, view=ft.WEB_BROWSER)

Error message

No response

------------------------------------------------------

• I have searched for answers to my question both in the issues and in previous discussions.

[ndonkoHenri]

Below is an AI suggestion. Let me know if it works (I haven't tried):

import tempfile
import os

# Assuming you have a blob containing audio data
audio_blob = b'...binary audio data...'  # Replace with your actual blob data

# Create a temporary file
with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as temp_audio_file:
    temp_audio_file.write(audio_blob)
    temp_audio_file_path = temp_audio_file.name

# Now use the temporary file path with your transcription library
# Example with a hypothetical transcription library
# Replace `transcription_lib.transcribe` with the actual function from your library

try:
    transcription_result = transcription_lib.transcribe(temp_audio_file_path)
    print("Transcription:", transcription_result)
finally:
    # Clean up the temporary file
    os.remove(temp_audio_file_path)

Which lib are you using? If it accepts in-memory file-like objects, you can try this:

# Create an in-memory file-like object from the blob
audio_file = io.BytesIO(audio_blob)

# Transcribe...

[leony777]

Hey, thanks a bunch for trying to help me out!
Still no luck, man. Even when I throw in the URL, the code just ain’t grabbing the audio data. I peeped that the memory gets freed up when I use that "audio_rec.start_recording_async(path)" thingy, but I can’t get my hands on the data; it just spits out some "coroutine" object.