qwen-audio-3.0-asr-flash
A speech recognition model that converts audio into readable text with punctuation and normalized numbers. Supports Russian and other languages, as well as Chinese dialects.
- Ailoot · Speech-to-text
POST /v1/audio/transcriptionsqwen-audio-3.0-asr-flash - Ailoot · Speech-to-text
POST /v1/audio/transcriptionsqwen-asr - Ailoot · Speech-to-text
POST /v1/audio/transcriptionsqwen-asr-latest
About the model
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Qwen-Audio-3.0-ASR-Flash is designed to convert recorded speech into text. Alibaba highlights multilingual recognition and handling of expressive speech, including poetic rhythm, classical Chinese and cultural storytelling. Audio is supplied as input, and a transcript is returned.
Speech in written form
The model inserts punctuation and converts spoken numbers, dates and monetary amounts into their conventional written forms. This is useful for voice messages, interview excerpts and educational speech where readable text is needed rather than just a sequence of recognized words.
It supports 30 languages, including Russian, English, French, German, Japanese and Arabic. It also accounts for seven major Chinese dialect groups and more than 20 regional pronunciation variants. This set is intended to recognize a range of language and accent characteristics.
Task context and hint words can be supplied for specialized terminology. They help guide recognition of names and domain-specific vocabulary. Timestamps are always enabled for this version, allowing the result to be linked to the corresponding sections of the recording. A short file's transcript may be returned in full or incrementally as processing proceeds.
Material and limitations
The maximum recording length is five minutes. One file is used per task. Supported formats are AAC, AMR, AVI, FLAC, FLV, M4A, MKV, MOV, MP3, MP4, MPEG, OGG, OPUS, WAV, WEBM, WMA and WMV; any sample rate is allowed. Video containers in this list refer to extracting speech from their audio tracks, not to visual analysis of their frames.
Qwen-Audio-3.0-ASR-Flash does not provide speaker diarization or emotion recognition. Its main output is text linked to the timing of the recognized content in the recording. Longer recording capabilities and other forms of audio analysis stated for neighboring models do not apply to this version.
Ailoot · Speech-to-text POST /v1/audio/transcriptions qwen-audio-3.0-asr-flash
qwen-audio-3.0-asr-flashqwen-audio-3.0-asr-flashqwen-asrqwen-asr-latest
Overview
A speech recognition model that converts audio into readable text with punctuation and normalized numbers. Supports Russian and other languages, as well as Chinese dialects.
- Accepted inputs
- Audio
- Results
- Text
Available in Ailoot · inputs & methods
Ailoot · Speech-to-text POST /v1/audio/transcriptions
- Input types
audio- Output types
text- Output format
- json
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen-audio-3.0-asr-flash | Default— |
Fieldfile | Typefile | RequirementRequired | Values & limits Maximum bytes: 10,485,760 File formats: wav, mp3, m4a, ogg, webm, flac Content type: multipart/form-data Maximum duration, seconds: 300 | Default— |
Fieldresponse_format | Typestring | RequirementOptional | Values & limits json | Default— |
Input material combinations
speech_to_text · Required: actual audio file
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contractaudio |
Parameterrequired | Contractnull |
Parameteroptional_controls | Contractnull |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/tasks/{task_id} · actual returned task_id or media library record
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
Method shapes and IDs are exact; supplier capability is separate from admission.
Pricing
Ailoot method rates
POST /v1/audio/transcriptions- Input
- $0.0021 per minute of audio
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Example requests
Ailoot · Speech-to-text
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
: "${AILOOT_AUDIO_FILE:?Set your actual MP3 recording}"
export AILOOT_AUDIO_FILE
python3 - <<'PY_REQUEST'
import os
from pathlib import Path
f=Path(os.environ["AILOOT_AUDIO_FILE"])
if not f.is_file() or f.suffix.lower()!=".mp3":raise SystemExit("Choose your actual MP3 recording")
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \
-H "Authorization: Bearer $AILOOT_API_KEY" \
--form-string model=qwen-audio-3.0-asr-flash -F "file=@$AILOOT_AUDIO_FILE" \
-D response-headers.txt -o result.json
python3 - <<'PY_RESULT'
import json
from pathlib import Path
print(json.dumps(json.loads(Path("result.json").read_text()),ensure_ascii=False,indent=2))
PY_RESULT
Receive and save results
Ailoot · Speech-to-text
json
GET /v1/media/tasks/{task_id}GET /v1/media/assets/{asset_id}/contentGET /v1/media/assetsGET /v1/media/assets/{asset_id}
Keep all completed results
Keep every complete returned response, file and content part.
Retain complete JSON, not only text; preserve every available segment/speaker/timestamp and actual task/asset identity.
If the original response needs checking
Retain original operation_id/X-Client-Request-ID for reconciliation without resubmission; no dedicated audio poll route.
Response fields
| Field | Type |
|---|---|
Fieldtext | Typestring |
Fieldsegments[] | Typeobject |
Fieldtask_id | Typestring |
Fieldasset_url | Typestring |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Responseserver_error | ConditionThe returned response says the operation or result requires a check | Next stepKeep the original returned operation/task ID, request headers and response; continue supported reads of that original operation. Do not resubmit the creation request. |
Aliases
Model aliases
Ailoot qwen-audio-3.0-asr-flash, qwen-asr, qwen-asr-latest
Method path aliases
Use the method paths shown above.
Related caller IDs
Ailoot · Speech-to-text POST /v1/audio/transcriptions qwen-asr
qwen-asrqwen-audio-3.0-asr-flashqwen-asrqwen-asr-latest
Overview
A speech recognition model that converts audio into readable text with punctuation and normalized numbers. Supports Russian and other languages, as well as Chinese dialects.
- Accepted inputs
- Audio
- Results
- Text
Available in Ailoot · inputs & methods
Ailoot · Speech-to-text POST /v1/audio/transcriptions
- Input types
audio- Output types
text- Output format
- json
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen-asr | Default— |
Fieldfile | Typefile | RequirementRequired | Values & limits Maximum bytes: 10,485,760 File formats: wav, mp3, m4a, ogg, webm, flac Content type: multipart/form-data Maximum duration, seconds: 300 | Default— |
Fieldresponse_format | Typestring | RequirementOptional | Values & limits json | Default— |
Input material combinations
speech_to_text · Required: actual audio file
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contractaudio |
Parameterrequired | Contractnull |
Parameteroptional_controls | Contractnull |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/tasks/{task_id} · actual returned task_id or media library record
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
Method shapes and IDs are exact; supplier capability is separate from admission.
Pricing
Ailoot method rates
POST /v1/audio/transcriptions- Input
- $0.0021 per minute of audio
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Example requests
Ailoot · Speech-to-text
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
: "${AILOOT_AUDIO_FILE:?Set your actual MP3 recording}"
export AILOOT_AUDIO_FILE
python3 - <<'PY_REQUEST'
import os
from pathlib import Path
f=Path(os.environ["AILOOT_AUDIO_FILE"])
if not f.is_file() or f.suffix.lower()!=".mp3":raise SystemExit("Choose your actual MP3 recording")
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \
-H "Authorization: Bearer $AILOOT_API_KEY" \
--form-string model=qwen-asr -F "file=@$AILOOT_AUDIO_FILE" \
-D response-headers.txt -o result.json
python3 - <<'PY_RESULT'
import json
from pathlib import Path
print(json.dumps(json.loads(Path("result.json").read_text()),ensure_ascii=False,indent=2))
PY_RESULT
Receive and save results
Ailoot · Speech-to-text
json
GET /v1/media/tasks/{task_id}GET /v1/media/assets/{asset_id}/contentGET /v1/media/assetsGET /v1/media/assets/{asset_id}
Keep all completed results
Keep every complete returned response, file and content part.
Retain complete JSON, not only text; preserve every available segment/speaker/timestamp and actual task/asset identity.
If the original response needs checking
Retain original operation_id/X-Client-Request-ID for reconciliation without resubmission; no dedicated audio poll route.
Response fields
| Field | Type |
|---|---|
Fieldtext | Typestring |
Fieldsegments[] | Typeobject |
Fieldtask_id | Typestring |
Fieldasset_url | Typestring |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Responseserver_error | ConditionThe returned response says the operation or result requires a check | Next stepKeep the original returned operation/task ID, request headers and response; continue supported reads of that original operation. Do not resubmit the creation request. |
Aliases
Model aliases
Ailoot qwen-audio-3.0-asr-flash, qwen-asr, qwen-asr-latest
Method path aliases
Use the method paths shown above.
Related caller IDs
Ailoot · Speech-to-text POST /v1/audio/transcriptions qwen-asr-latest
qwen-asr-latestqwen-audio-3.0-asr-flashqwen-asrqwen-asr-latest
Overview
A speech recognition model that converts audio into readable text with punctuation and normalized numbers. Supports Russian and other languages, as well as Chinese dialects.
- Accepted inputs
- Audio
- Results
- Text
Available in Ailoot · inputs & methods
Ailoot · Speech-to-text POST /v1/audio/transcriptions
- Input types
audio- Output types
text- Output format
- json
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen-asr-latest | Default— |
Fieldfile | Typefile | RequirementRequired | Values & limits Maximum bytes: 10,485,760 File formats: wav, mp3, m4a, ogg, webm, flac Content type: multipart/form-data Maximum duration, seconds: 300 | Default— |
Fieldresponse_format | Typestring | RequirementOptional | Values & limits json | Default— |
Input material combinations
speech_to_text · Required: actual audio file
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contractaudio |
Parameterrequired | Contractnull |
Parameteroptional_controls | Contractnull |
Lifecycle: synchronous_result_then_owned_library
GET /v1/media/tasks/{task_id} · actual returned task_id or media library record
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}
Method shapes and IDs are exact; supplier capability is separate from admission.
Pricing
Ailoot method rates
POST /v1/audio/transcriptions- Input
- $0.0021 per minute of audio
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Example requests
Ailoot · Speech-to-text
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
: "${AILOOT_AUDIO_FILE:?Set your actual MP3 recording}"
export AILOOT_AUDIO_FILE
python3 - <<'PY_REQUEST'
import os
from pathlib import Path
f=Path(os.environ["AILOOT_AUDIO_FILE"])
if not f.is_file() or f.suffix.lower()!=".mp3":raise SystemExit("Choose your actual MP3 recording")
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \
-H "Authorization: Bearer $AILOOT_API_KEY" \
--form-string model=qwen-asr-latest -F "file=@$AILOOT_AUDIO_FILE" \
-D response-headers.txt -o result.json
python3 - <<'PY_RESULT'
import json
from pathlib import Path
print(json.dumps(json.loads(Path("result.json").read_text()),ensure_ascii=False,indent=2))
PY_RESULT
Receive and save results
Ailoot · Speech-to-text
json
GET /v1/media/tasks/{task_id}GET /v1/media/assets/{asset_id}/contentGET /v1/media/assetsGET /v1/media/assets/{asset_id}
Keep all completed results
Keep every complete returned response, file and content part.
Retain complete JSON, not only text; preserve every available segment/speaker/timestamp and actual task/asset identity.
If the original response needs checking
Retain original operation_id/X-Client-Request-ID for reconciliation without resubmission; no dedicated audio poll route.
Response fields
| Field | Type |
|---|---|
Fieldtext | Typestring |
Fieldsegments[] | Typeobject |
Fieldtask_id | Typestring |
Fieldasset_url | Typestring |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Responseserver_error | ConditionThe returned response says the operation or result requires a check | Next stepKeep the original returned operation/task ID, request headers and response; continue supported reads of that original operation. Do not resubmit the creation request. |
Aliases
Model aliases
Ailoot qwen-audio-3.0-asr-flash, qwen-asr, qwen-asr-latest
Method path aliases
Use the method paths shown above.
Related caller IDs
FAQ · qwen-audio-3.0-asr-flash
- What inputs and results are available?
POST /v1/audio/transcriptions·qwen-audio-3.0-asr-flash:audio→textPOST /v1/audio/transcriptions·qwen-asr:audio→textPOST /v1/audio/transcriptions·qwen-asr-latest:audio→text- Which fields does the request use?
POST /v1/audio/transcriptions·qwen-audio-3.0-asr-flash:model,file,response_formatPOST /v1/audio/transcriptions·qwen-asr:model,file,response_formatPOST /v1/audio/transcriptions·qwen-asr-latest:model,file,response_format