# ------------------------------------ # Copyright (c) Microsoft Corporation. # Licensed under the MIT License. # ------------------------------------ import os import pytest import pathlib import uuid from devtools_testutils import AzureRecordedTestCase from conftest import WHISPER_AZURE, OPENAI, PREVIEW, GA, configure, TTS_OPENAI, TTS_AZURE audio_test_file = pathlib.Path(__file__).parent / "./assets/hello.m4a" audio_long_test_file = pathlib.Path(__file__).parent / "./assets/wikipediaOcelot.wav" @pytest.mark.live_test_only class TestAudio(AzureRecordedTestCase): @configure @pytest.mark.parametrize( "api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")] ) def test_transcribe(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_test_file, "rb"), **kwargs, ) assert result.text == "Hello." @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_raw(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( # file=open(audio_test_file, "rb").read(), # use file tuple for now file=("hello.m4a", open(audio_test_file, "rb").read(), "application/octet-stream"), **kwargs, ) assert result.text == "Hello." @configure @pytest.mark.parametrize( "api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")] ) def test_translate(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_test_file, "rb"), **kwargs, ) assert result.text == "Hello." @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_raw(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( # file=open(audio_test_file, "rb").read(), # use file tuple for now file=("hello.m4a", open(audio_test_file, "rb").read(), "application/octet-stream"), **kwargs, ) assert result.text == "Hello." @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_verbose(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_long_test_file, "rb"), response_format="verbose_json", **kwargs, ) assert result.text == "The ocelot, Lepardus paradalis, is a small wild cat native to the southwestern " \ "United States, Mexico, and Central and South America. This medium-sized cat is characterized by " \ "solid black spots and streaks on its coat, round ears, and white neck and undersides. It weighs " \ "between 8 and 15.5 kilograms, 18 and 34 pounds, and reaches 40 to 50 centimeters – 16 to 20 inches " \ "– at the shoulders. It was first described by Carl Linnaeus in 1758. Two subspecies are recognized, " \ "L. p. paradalis and L. p. mitis. Typically active during twilight and at night, the ocelot tends to " \ "be solitary and territorial. It is efficient at climbing, leaping, and swimming. It preys on small " \ "terrestrial mammals such as armadillo, opossum, and lagomorphs." assert result.task == "transcribe" assert result.language == "english" assert result.duration == 56.25 for segment in result.segments: assert segment.id is not None assert segment.seek is not None assert segment.start is not None assert segment.end is not None assert segment.text is not None assert segment.tokens is not None assert segment.temperature is not None assert segment.avg_logprob is not None assert segment.compression_ratio is not None assert segment.no_speech_prob is not None @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_text(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_test_file, "rb"), response_format="text", **kwargs, ) assert result == "Hello.\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_srt(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_test_file, "rb"), response_format="srt", **kwargs, ) assert result == "1\n00:00:00,000 --> 00:00:02,000\nHello.\n\n\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_vtt(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_test_file, "rb"), response_format="vtt", **kwargs, ) assert result == "WEBVTT\n\n00:00:00.000 --> 00:00:02.000\nHello.\n\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_verbose(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_long_test_file, "rb"), response_format="verbose_json", **kwargs, ) assert result.text == "The ocelot, Lepardus paradalis, is a small wild cat native to the southwestern " \ "United States, Mexico, and Central and South America. This medium-sized cat is characterized by " \ "solid black spots and streaks on its coat, round ears, and white neck and undersides. It weighs " \ "between 8 and 15.5 kilograms, 18 and 34 pounds, and reaches 40 to 50 centimeters – 16 to 20 inches " \ "– at the shoulders. It was first described by Carl Linnaeus in 1758. Two subspecies are recognized, " \ "L. p. paradalis and L. p. mitis. Typically active during twilight and at night, the ocelot tends to " \ "be solitary and territorial. It is efficient at climbing, leaping, and swimming. It preys on small " \ "terrestrial mammals such as armadillo, opossum, and lagomorphs." assert result.task == "translate" assert result.language == "english" assert result.duration == 56.25 for segment in result.segments: assert segment.id is not None assert segment.seek is not None assert segment.start is not None assert segment.end is not None assert segment.text is not None assert segment.tokens is not None assert segment.temperature is not None assert segment.avg_logprob is not None assert segment.compression_ratio is not None assert segment.no_speech_prob is not None @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_text(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_test_file, "rb"), response_format="text", **kwargs, ) assert result == "Hello.\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_srt(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_test_file, "rb"), response_format="srt", **kwargs, ) assert result == "1\n00:00:00,000 --> 00:00:02,000\nHello.\n\n\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_vtt(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_test_file, "rb"), response_format="vtt", **kwargs, ) assert result == "WEBVTT\n\n00:00:00.000 --> 00:00:02.000\nHello.\n\n" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_transcribe_options(self, client, api_type, api_version, **kwargs): result = client.audio.transcriptions.create( file=open(audio_test_file, "rb"), temperature=0, language="en", prompt="Transcribe the text exactly as spoken.", **kwargs, ) assert result.text == "Hello" @configure @pytest.mark.parametrize("api_type, api_version", [(WHISPER_AZURE, GA), (WHISPER_AZURE, PREVIEW), (OPENAI, "v1")]) def test_translate_options(self, client, api_type, api_version, **kwargs): result = client.audio.translations.create( file=open(audio_test_file, "rb"), temperature=0, prompt="Translate the text exactly as spoken.", **kwargs, ) assert result.text == "Hello" @configure @pytest.mark.parametrize( "api_type, api_version", [(TTS_AZURE, PREVIEW), (TTS_OPENAI, "v1")] ) def test_tts(self, client, api_type, api_version, **kwargs): speech_file_path = pathlib.Path(__file__).parent / f"{uuid.uuid4()}.mp3" try: response = client.audio.speech.create( voice="alloy", input="The quick brown fox jumped over the lazy dog.", **kwargs, ) assert response.encoding assert response.content assert response.text response.write_to_file(speech_file_path) finally: os.remove(speech_file_path) @configure @pytest.mark.parametrize("api_type, api_version", [(TTS_AZURE, PREVIEW), (TTS_OPENAI, "v1")]) def test_tts_hd_streaming(self, client, api_type, api_version, **kwargs): with client.audio.speech.with_streaming_response.create( voice="echo", input="The quick brown fox jumped over the lazy dog.", model="tts-1-hd" ) as response: response.read() @configure @pytest.mark.parametrize("api_type, api_version", [(TTS_AZURE, PREVIEW), (TTS_OPENAI, "v1")]) def test_tts_response_format(self, client, api_type, api_version, **kwargs): speech_file_path = pathlib.Path(__file__).parent / f"{uuid.uuid4()}.flac" try: response = client.audio.speech.create( voice="fable", input="The quick brown fox jumped over the lazy dog.", response_format="flac", **kwargs ) assert response.encoding assert response.content assert response.text response.stream_to_file(speech_file_path) # deprecated finally: os.remove(speech_file_path) @configure @pytest.mark.parametrize("api_type, api_version", [(TTS_AZURE, PREVIEW), (TTS_OPENAI, "v1")]) def test_tts_speed(self, client, api_type, api_version, **kwargs): speech_file_path = pathlib.Path(__file__).parent / f"{uuid.uuid4()}.mp3" try: response = client.audio.speech.create( voice="onyx", input="The quick brown fox jumped over the lazy dog.", speed=3.0, **kwargs ) assert response.encoding assert response.content assert response.text response.write_to_file(speech_file_path) finally: os.remove(speech_file_path)