Skip to main content

Python Tinkoff Speech API

Project description

Tinkoff Python Speech API examples

Usage

Install from PyPi

pip install tinkoff-voicekit-client

Common

Return documentation for public methods

client.something_method.__doc__

Methods initialize using config (Python dict) which satisfies one of the next json schema.

Recogniton (STT)

Base types schema:

definitions = {
        "StringArray": {
            "type": "array",
            "items": {
                "type": "string",
            }
        },
        "AudioEncoding": {
            "type": "string",
            "enum": ["LINEAR16", "ALAW", "MULAW", "LINEAR32F", "RAW_OPUS", "MPEG_AUDIO"]
        },
        "VoiceActivityDetectionConfig": {
            "type": "object",
            "properties": {
                "min_speech_duration": {"type": "number"},
                "max_speech_duration": {"type": "number"},
                "silence_duration_threshold": {"type": "number"},
                "silence_prob_threshold": {"type": "number"},
                "aggressiveness": {"type": "number"},
            }
        },
        "SpeechContext": {
            "type": "object",
            "properties": {
                "phrases": {"$ref": "#definitions/StringArray"},
                "words": {"$ref": "#definitions/StringArray"}
            }
        },
        "InterimResultsConfig": {
            "type": "object",
            "properties": {
                "enable_interim_results": {"type": "boolean"},
                "interval": {"type": "number"}
            }
        }
    }

Recognition config schema:

recognition_config_schema = {
        "type": "object",
        "definitions": definitions,
        "properties": {
            "encoding": {"$ref": "#/definitions/AudioEncoding"},
            "sample_rate_hertz": {"type": "number"},
            "language_code": {"type": "string"},
            "max_alternatives": {"type": "number"},
            "speech_contexts": {
                "type": "array",
                "items": {
                    "$ref": "#/definitions/SpeechContext"
                }
            },
            "enable_automatic_punctuation": {"type": "boolean"},
            "model": {"type": "string"},
            "num_channels": {"type": "number"},
            "do_not_perform_vad": {"type": "boolean"},
            "vad_config": {"$ref": "#/definitions/VoiceActivityDetectionConfig"}
        },
        "required": [
            "sample_rate_hertz",
            "num_channels",
            "encoding",
        ],
        "additionalProperties": False
    }

Streaming recognition config schema:

streaming_recognition_config_schema = {
        "type": "object",
        "definitions": definitions,
        "properties": {
            "config": recognition_config_schema,
            "single_utterance": {"type": "boolean"},
            "interim_results_config": {"$ref": "#/definitions/InterimResultsConfig"}
        },
        "additionalProperties": False
    }

Example of using STT

from tinkoff_voicekit_client import ClientSTT

API_KEY = "my_api_key"
SECRET_KEY = "my_secret_key"

client = ClientSTT(API_KEY, SECRET_KEY)

my_config = {
    "encoding": "LINEAR16",
    "sample_rate_hertz": 8000,
    "num_channels": 1
}

# recognise method call
response = client.recognize("path/to/audio/file", my_config)
print(response)
from tinkoff_voicekit_client import ClientSTT

API_KEY = "my_api_key"
SECRET_KEY = "my_secret_key"

client = ClientSTT(API_KEY, SECRET_KEY)

my_config = {
    "encoding": "LINEAR16",
    "sample_rate_hertz": 8000,
    "num_channels": 1
}
stream_config = {"config": my_config}

# recognise stream method call
with open("path/to/audio/file", "rb") as source:
    responses = client.streaming_recognize(source, stream_config)
    for response in responses:
        print(response)

Example of Voice Activity Detection configuration

vad = {}
vad["min_speech_duration"] = min_speech_duration
vad["max_speech_duration"] = max_speech_duration
vad["silence_duration_threshold"] = silence_duration_threshold
vad["silence_prob_threshold"] = silence_prob_threshold
vad["aggressiveness"] = aggressiveness

my_config = {}
my_config["vad"] = vad

Synthesize (TTS)

Base types schema:

definitions = {
        "AudioEncoding": {
            "type": "string",
            "enum": ["LINEAR16", "ALAW", "MULAW", "LINEAR32F", "RAW_OPUS"]
        },
        "SynthesisInput": {
            "type": "object",
            "properties": {
                "text": {"type": "string"}
            }
        }
    }

Streaming synthesis schema:

streaming_synthesize_config_schema = {
        "type": "object",
        "definitions": definitions,
        "properties": {
            "audio_encoding": {"$ref": "#/definitions/AudioEncoding"},
            "speaking_rate": {"type": "number"},
            "sample_rate_hertz": {"type": "number"}
        },
        "required": [
            "sample_rate_hertz",
            "audio_encoding",
        ],
        "additionalProperties": False
    }

Example of input file:

Я жду вашего ответа. Вы готовы сделать перевод?
# Давайте уточним получателя. Как его зовут?

commented lines # will not be synthesis

Example of using TTS

from tinkoff_voicekit_client import ClientTTS

API_KEY = "api_key"
SECRET_KEY = "secret_key"

client = ClientTTS(API_KEY, SECRET_KEY)
my_config = {
    "audio_encoding": "LINEAR16",
    "sample_rate_hertz": 48000
}


# use it if you want work with proto results
# audio file
rows_responses = client.streaming_synthesize("path/to/file/with/text", my_config)
# text
rows_responses = client.streaming_synthesize("Мой красивый текст", my_config)

# use it if you want get audio file results
# audio file
client.synthesize_to_audio_wav("path/to/file/with/text", my_config, "output/dir")
# text
client.synthesize_to_audio_wav("Мой красивый текст", my_config, "output/dir")

Project details


Release history Release notifications

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for tinkoff-voicekit-client, version 0.1.1
Filename, size File type Python version Upload date Hashes
Filename, size tinkoff_voicekit_client-0.1.1-py3-none-any.whl (1.1 MB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size tinkoff_voicekit_client-0.1.1.tar.gz (1.1 MB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page