Audio

# Convert user query to text, process, respond with speech
with open("user_query.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(
        model="groq/whisper-large-v3-turbo",  # Fast
        file=audio
    )

# Process with LLM
response = client.chat.completions.create(
    model="openai/gpt-5-mini",
    messages=[{"role": "user", "content": transcript.text}]
)

# Convert response to speech
audio = client.audio.speech.create(
    model="openai/tts-1",
    voice="nova",
    input=response.choices[0].message.content
)
audio.stream_to_file("response.mp3")

with open("meeting.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(
        model="openai/whisper-1",
        file=audio,
        response_format="verbose_json",
        timestamp_granularities=["segment"]
    )

# Process segments for speaker diarization
for segment in transcript.segments:
    print(f"[{segment['start']:.2f}s - {segment['end']:.2f}s]: {segment['text']}")

# Translate foreign language audio to English
with open("spanish_content.mp3", "rb") as audio:
    translation = client.audio.translations.create(
        model="openai/whisper-1",
        file=audio,
        response_format="srt"  # Get subtitles
    )

with open("english_subtitles.srt", "w") as f:
    f.write(translation)

# Generate audio description for visually impaired users
description = "A serene mountain landscape at sunset..."

audio = client.audio.speech.create(
    model="openai/tts-1-hd",  # High quality
    voice="shimmer",  # Gentle voice
    input=description,
    speed=0.9  # Slightly slower for clarity
)
audio.stream_to_file("description.mp3")