tests/test_whisperer.py

import streamlit as st
import requests
import sounddevice as sd
import numpy as np
import tempfile
import wave
import os

# Hugging Face endpoint
HF_ENDPOINT = "https://lngcv623mxnv20nt.eu-west-1.aws.endpoints.huggingface.cloud"
HF_TOKEN = os.getenv("HF_TOKEN")  # Or replace with your token string

# Function to record audio
def record_audio(duration=5, samplerate=16000):
    st.info("Recording... Speak now 🎙️")
    audio = sd.rec(int(duration * samplerate), samplerate=samplerate, channels=1, dtype='int16')
    sd.wait()
    return audio, samplerate

# Save audio to temp WAV file
def save_wav(audio, samplerate):
    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".wav")
    with wave.open(temp_file.name, 'wb') as wf:
        wf.setnchannels(1)
        wf.setsampwidth(2)  # 16-bit
        wf.setframerate(samplerate)
        wf.writeframes(audio.tobytes())
    return temp_file.name

# Send audio to Hugging Face endpoint
def transcribe(audio_path):
    with open(audio_path, "rb") as f:
        response = requests.post(
            HF_ENDPOINT,
            headers={
                "Authorization": f"Bearer {HF_TOKEN}",
                "Content-Type": "audio/wav"
            },
            data=f
        )
    if response.status_code == 200:
        return response.json()
    else:
        st.error(f"Error: {response.status_code} - {response.text}")
        return None

# Streamlit UI
st.title("🎤 Whisper Tiny Transcriber")

duration = st.slider("Recording duration (seconds):", 3, 15, 5)

if st.button("Press to Record"):
    audio, sr = record_audio(duration)
    wav_path = save_wav(audio, sr)

    st.audio(wav_path, format="audio/wav")

    result = transcribe(wav_path)
    if result:
        st.subheader("Transcription:")
        st.write(result.get("text", result))